Knowledge Base Question Generation (KBQG) aims to generate human-readable questions from a collection of factual triples and a given answer entity. Approaches that utilize pre-trained language models improve question generation performance, yet they are still limited by the inherent limitations of these models. In contrast, Large Language Models (LLMs) possess more advanced capabilities in language understanding and generation. However, directly applying LLMs often results in questions that do not correspond well to the target answers or the structure of the subgraph. To address this issue, we devise DERA, a Dual-Encoder Retrieval-Augmented framework that guides LLMs using structurally similar subgraph examples as enhanced prompts. DERA identifies similar subgraphs through a dual encoder that captures structural features from multiple perspectives and applies contrastive learning to optimize the representations. Experiments on the WebQuestions and PathQuestions datasets show that DERA achieves consistent improvements across multiple evaluation metrics and shows strong cross-domain adaptability.

Group Activity Recognition (GAR) aims to infer collective behaviors from the motions and interactions of multiple people in a scene. Compared with RGB-based pipelines, skeleton-based methods are more efficient and less sensitive to background appearance, but they usually rely on fixed graph topologies and are vulnerable to pose estimation noise. These two issues are especially harmful in crowded sports scenes, where interaction patterns change rapidly and upstream pose estimators may produce missing, swapped, or jittered joints. To address this problem, we propose a condensed presentation of our Contrastive Memory-Augmented Adaptive Graph Convolutional Network (CMAA-GCN). The framework is built on two key ideas. First, a Dual-Level Adaptive Spatial Graph Convolution (DASGC) module separates joint-level motion modeling from group-level interaction modeling and learns adaptive adjacency matrices for both levels. Second, an error-aware contrastive supervision mechanism simulates realistic pose failures and uses dual memory banks to align noisy and clean representations. This design improves both adaptability and robustness. Experiments on the reannotated Volleyball dataset, the original Volleyball dataset, the NBA benchmark, and Kinetics-400 show that the proposed method achieves strong accuracy while preserving the efficiency advantage of skeleton-based recognition. The model reaches 97.8% on reannotated Volleyball, 96.2% on original Volleyball, 76.1% on NBA, and 50.6% on Kinetics-400.

Constructing high-fidelity document-level event graphs remains a significant challenge for Large Language Models (LLMs). Existing generative paradigms predominantly model extracted events as discrete symbolic nodes, an assumption that falters in complex real-world documents. Specifically, conventional symmetric similarity metrics fail to align sparse abstract mentions with information-dense backbone events, exacerbating macro-structural fragmentation. Concurrently, mitigating generative hallucinations via rigid binary thresholding severely compromises global topological connectivity. To overcome these structural and logical bottlenecks, we propose a Fuzzy-Enhanced Physical-Logical Synergistic Framework. On the physical front, we project discrete extractions into a continuous semantic space via Probabilistic Semantic Mass (PSM), a novel density-aware representation. This allows high-mass backbone events to dynamically absorb scattered fragments through asymmetric gravitation, bypassing the penalties of symmetric matching. On the logical front, we introduce a Fuzzy Logic Reasoner that replaces hard graph truncation with continuous soft penalties, effectively suppressing temporal and causal hallucinations without severing valid multi-hop reasoning chains. Extensive experiments on ChFinAnn, WikiEvents, and EventStoryLine demonstrate that our framework substantially outperforms state-of-the-art baselines. By shifting from discrete symbolic mapping to continuous physical-logical synergy, our approach establishes a highly robust paradigm for document-level structured event extraction.

Real bank-transaction data are rarely available to non-regulatory parties due to privacy and compliance constraints, while public substitutes often mismatch financial semantics, limiting the transferability and reproducibility of anomaly-detection research in financial settings. To address this, we release F2-Gen: an open-source, scenario-driven, configurable web platform that generates synthetic data covering both tabular transaction records and transaction graphs for modeling and evaluating financial-fraud behaviors. F2-Gen provides parameterized fraud scenarios, interactive visual analysis, automated screening to remove label-revealing fields, tutorials, and batch data export. Use cases include anomaly detection, graph learning, and teaching demonstrations; standard benchmarking is supported via a screened feature subset, ready-to-run baseline scripts, and sanity/controllability checks. Project page: https://sethgu.github.io/FinancialFraudDataGenerator/. Code: https://github.com/sethGu/FinancialFraudDataGenerator.

Unsupervised graph anomaly detection seeks to discover rare patterns in graph-structured data, enabling applications across diverse fields such as financial transactions, e-commerce, and beyond. Although recent methods based on graph autoencoder and graph contrastive consistency show considerable promise in this field, the former primarily addresses global anomalies, whereas the latter is more focused on local anomalies. This distinction limits their capacity to fully capture the abnormalities of nodes in the graph. Moreover, challenges such as overfitting of abnormal patterns and the introduction of unknown noise from random walk-based subgraph sampling remain prevalent. These issues contribute to suboptimal model performance. To tackle these challenges, we propose Local and Global Contrastive Networks (LGCN). LGCN first extracts graph features using a feature encoder module and subsequently projects them into local and global feature subspaces. The global information learning module mitigates overfitting in reconstruction techniques to abnormal patterns by simultaneously accounting for both the similarities and differences between nodes. The local information learning module extracts anomaly information by evaluating the consistency between the center node and its corresponding context features. Extensive experiments across eight datasets demonstrate that LGCN outperforms existing methods in unsupervised graph anomaly detection.

Knowledge Base Question Answering (KBQA) aims to answer natural language questions by reasoning over large-scale structured Knowledge Bases (KBs). Among existing approaches, semantic parsing-based methods have emerged as a mainstream solution, where Large Language Models (LLMs) are employed to translate questions into structured graph queries such as Logical Forms (LFs). However, this paradigm faces two critical challenges: (1) The complex semantic mapping and graph retrieval operations render direct one-shot LF generation difficult; (2) LLMs suffer from inherent hallucination issues, generating semantically plausible-seeming but factually incorrect or invalid LFs, which are non-executable. To address these challenges, this article proposes HP-Corr , a novel framework that integrates H op-wise P lanning with iterative explainable self- Corr ection for faithful knowledge reasoning. Specifically, the system utilizes a fine-tuned open source LLM for query planning and explainable self-correction. The query planner generates reasoning paths hop-by-hop, while an explainable self-correction provides hop-wise feedback, enabling interpretable path editing based on existing reasoning paths and retrieved KB knowledge. By introducing the dual-module cooperative architecture, our system performs iterative plan-then-correct to refine query paths progressively, ensuring answer reliability and LFs executability. Experimental results demonstrate significant improvements, with our approach achieving higher accuracy while substantially reducing the search space, particularly in complex multi-hop KBQA scenarios.

Graph-based fraud detection is crucial in identifying illegal activities in social networks, finance, and other sectors. Despite recent progress in this area, most of current researches typically require a large amount of annotated data to demonstrate its benefits. In practice, obtaining sufficient high-quality annotated data is challenging, limiting the effectiveness of model training. Therefore, leveraging extremely limited label information is crucial to enhance model performance. We propose a context-aware graph neural network (CGNN) to address this. CGNN performs category semantic decomposition on the contextual neighbor features of the center node to enrich the category semantics. In the neighbor message aggregation stage, the denoising attention mechanism enables the center node to adaptively aggregate heterophilic and homophilic information from neighbors. Particularly for unlabeled data, feature augmentation within the category subspace and consistency regularization driven by entropy minimization ensure that such data can further enhance model performance under explicit semantic guidance. We demonstrate on four real-world datasets that CGNN significantly outperforms other baseline methods with extremely limited labels.

In UAV-UGV collaboration, UGVs enhance their perception capabilities by utilizing shared observation data from UAVs to make effective decisions. However, learning navigation policies directly from multi-perspective observation data is challenging due to spatial inconsistencies and redundancy. Existing approaches either rely on low-level scene representations, which may reduce efficiency, or use mid-level semantic representations that lack essential geometric and temporal details, resulting in sub-optimal decisions. This paper presents STSR-Nav, a novel framework for navigation policy learning based on UAV-UGV collaborative perception. The framework first encodes semantic information and geometric relationships from UAV images using a Bird'View (BEV) scene graph. A learnable graph neural network then integrates the BEV scene graph with local perception to infer spatial features. Temporal learning is applied to these features via a memory network, generating spatial-temporal features that are embedded into the UGV's local perception space using a multi-head attention mechanism. Finally, the fused spatial-temporal features provide a high-level scene representation for navigation policy learning. Unity simulations demonstrate the framework's improved efficiency and generalization capabilities.

The large-scale knowledge base question-answering (KBQA) has become increasingly vital across various fields. In the era of large language models (LLMs), leveraging knowledge base retrieval combined with large models for knowledge reasoning has become the mainstream approach for KBQA. However, this method faces two primary challenges: (1) the high computational cost and low accuracy of similarity-based path retrieval, and (2) the relatively low accuracy of directly obtaining answers from large models. In this paper, we introduce a novel method of retrieval enhancement-stepwise reasoning (RESR), which transforms path retrieval into text semantic understanding to minimize unnecessary interference from path information in the reasoning process, guiding the LLM to generate interpretable reasoning paths rather than directly producing answers. Specifically, RESR fine-tunes the generative model through text semantic understanding to swiftly and accurately filter path information relevant to the query from a large-scale knowledge graph (KG). Additionally, we employ the Chain-of-Thought (CoT) method to guide LLMs in step-by-step reasoning, verifying the logical coherence of reasoning paths rather than directly deriving answers. Our proposed method achieves state-of-the-art (SOTA) performance on the WebQuestionsSP (WQSP) and ComplexWebQuestions (CWQ) benchmarks.

In graph-based fraud detection, accurately distinguishing between normal and anomalous account nodes is crucial. Recent graph-based approaches have achieved remarkable progress, primarily by aggregating neighborhood information to learn node representations and enhance model discriminability. However, most existing methods employ a unified aggregation function to neighbors with different class labels, which weakens feature differences and allows anomalous node features to be obscured by normal ones, leading to representation contamination. Moreover, existing methods assign uniform confidence weights to all labeled nodes during training, without accounting for label quality differences caused by varying distances to the decision boundary, which weakens the robustness of model learning. To overcome these limitations, we propose a novel method named Confidence-Aware Neighborhood Aggregation (CANA). CANA introduces a category-aware aggregation mechanism that performs separate aggregation operations for normal, anomalous, and unlabeled neighbors, thereby mitigating the issue of representation contamination. In addition, CANA incorporates a confidence-aware node-reweighted learning objective function, which dynamically adjusts the confidence weights of the labeled nodes based on a confidence metric computed through controlled perturbations in neighbors and prediction accuracy, thus improving the model's ability to leverage high quality labeled nodes and improving its learning capability and classification performance. Specifically, the function includes two optional perturbation strategies, namely Homo-Hetero Neighbor Perturbation (HHNP) and Robust Neighbor Perturbation (RNP). Extensive evaluations of four public benchmark datasets underscore the superiority of CANA over current state-of-the-art methods.

Knowledge graph-to-text (KG-to-text) generation involves generating fluent and faithful text from structured data within knowledge graphs. Existing methods typically linearize the graph data and use pre-trained models, but they often neglect the complexity of graph structures and relations, limiting their ability to capture information effectively. In this paper, we propose a hierarchical attention model named ESSI-KG, which utilizes positional embeddings and attention mechanisms to enhance the representation and learning of relational information within knowledge graphs. We design three types of positional visualization embedding layers to encode graph structure information into sequences, thereby more effectively capturing the local information within triples in the knowledge graph. Additionally, we introduce entity attention layers and relation attention layers to retain global graph structure information and focus on relations among triples in the knowledge graph. Extensive experiments conducted on four benchmark datasets provide compelling evidence that ESSI-KG outperforms existing state-of-the-art models, demonstrating enhanced logical consistency and producing more accurate and informative textual outputs. Our model attains a BLEU-4 score that exceeds other baselines by over 1.27 on WebNLG. Specifically, under the few-shot settings, ESSI-KG reaches baseline performance with only one-twentieth of the labeled examples typically required.

Event Causality Identification (ECI) aims to determine whether a causal relationship exists between two events within a document. Most existing approaches rely on pre-constructed event graphs to capture event relationships. However, these methods often neglect the semantic context of events in the original document and struggle to model complex relationships using a single event graph. We propose that event relationship identification should progress from simpler to more complex relationships, with a strong reliance on global semantic information. To address these challenges, we propose an Iterative Progressive Reasoning Network for ECI. First, we introduce an Event Semantic-Aware Module, which captures the global semantic context of events in the document. Then, we design an Iterative Attention Graph Fusion Module, inspired by Bayesian reasoning, that dynamically generates event association graphs to progressively refine event relationships. This approach enables the model to perform high-order reasoning while maintaining semantic coherence across the document. Additionally, a Differential Regularizer constrains the receptive fields of both the semantic-aware and reasoning modules, ensuring they focus on distinct aspects of event relationships. Our experimental results demonstrate that the proposed method outperforms existing state-of-the-art approaches on two benchmark datasets, EventStoryLine and Causal-TimeBank.

Fine-tuning pre-trained language models (PLMs) for downstream tasks has achieved remarkable success in various natural language processing (NLP) applications. To further enhance the overall performance of PLMs across different NLP tasks, we propose a novel adversarial contrastive training (ACT) method that incorporates adversarial perturbations in the parameter space. Specifically, ACT introduces adversarial perturbations to the model’s parameters, deliberately degrading its performance on downstream tasks. Subsequently, we apply contrastive learning to align the representations of specific tasks with those of the perturbed model, thereby improving the model’s robustness and generalization ability. Experimental results demonstrate that ACT significantly enhances the generalization performance of PLMs and achieves state-of-the-art results on several text classification benchmarks.

Heterogeneous multi-agent reinforcement learning algorithms have been introduced to tackle complex real-world tasks such as gaming, disaster rescue, and navigation. The ability of heterogeneous agents to collaborate effectively is crucial for solving these complex tasks. However, the vast joint action and state space makes achieving cooperation among heterogeneous agents particularly challenging. Existing heterogeneous agent algorithms struggle to facilitate effective cooperation, and current role-based methods are not applicable due to the differing observation and action dimensions of heterogeneous agents. This paper proposes a novel algorithm called the Heterogeneous Agent Role Function Tree (HARFT), which forms corresponding roles based on the different functions of heterogeneous agents, allowing them to cooperate effectively. Additionally, we introduce the Heterogeneous Agent Role Predictive Network (HARPN), which enhances cooperation by predicting the current roles of other agents and adjusting collaboration based on their different functions. Finally, the effectiveness of our algorithm is validated in both the SMAC environment and a heterogeneous agent cooperative environment.

Unsupervised graph anomaly detection methods have attracted significant research interest in recent years. However, existing graph neural network models face a key challenge: they unintentionally integrate the behavioral features of normal nodes into the features of abnormal nodes during the aggregation process. This problem misleads model training, weakens the generalization, and complicates the anomaly detection. To address this challenge, we propose a Feature Fusion Contrastive Learning and Adaptive Correction framework, called FFCAC. The framework introduces a multi-scale feature fusion and contrastive learning approach that decomposes raw node features into multiple subspaces and applies multi-scale convolution to capture higher-order behavioral representations of nodes. Through contrastive learning, FFCAC computes the anomaly scores of node features, replacing the potentially misleading message passing mechanism commonly used in traditional GNNs. In addition, a Transformer-based correction mechanism is used to refine the anomaly scores by modelling global structural relationships through learned weights. This design is able to adaptively aggregate neighbourhood information, which enhances the generalization ability of the model while effectively identifying camouflaged abnormal nodes. Experimental results on multiple open-source datasets demonstrate that FFCAC achieves significant improvements in unsupervised graph anomaly detection compared to existing methods.

Graph anomaly detection plays a crucial role in identifying nodes that deviate significantly from normal patterns within a graph, with applications spanning various domains such as detection of authorship fraud and rumor propagation. Traditional methods primarily focus on aggregating information from neighboring nodes and reconstructing the central node based on these aggregated features. The anomaly degree is then calculated by comparing the reconstructed features with the original ones. Despite their effectiveness, these methods face limitations due to the constraints of device performance and the need to protect user privacy. In reality, graph data is often partitioned and distributed across different local clients, which leads to isolated client subgraphs. This partitioning results in incomplete feature aggregation, as the connections between subgraphs are missing, ultimately reducing the performance of anomaly detection models. To overcome these challenges, a federated graph anomaly detection approach based on disentangled representation learning is proposed. This method separates node features into two distinct components: intrinsic features and subgraph style features. By identifying outliers within the subgraph style features, a set of pseudo-nodes is generated and shared across the entire graph. These pseudo-nodes simulate connections between otherwise isolated subgraphs, which enables more comprehensive aggregation of intrinsic features from neighboring nodes. In addition, conditional variational autoencoders (CVAE) are employed alongside contrastive learning strategies to alleviate class imbalance and achieve effective feature disentanglement. These techniques help ensure that anomalous nodes are detected more accurately despite the inherent challenges of federated graph systems. Extensive experiments conducted on six diverse datasets provide compelling evidence of the proposed method's superior performance in federated graph anomaly detection, highlighting its ability to effectively handle incomplete graph structures while maintaining data privacy.

ABSTRACT Unmanned aerial vehicles (UAVs) are increasingly vital across numerous sectors, from logistics and rescue operations to military endeavours and beyond. However, ensuring safety in the decision‐making processes surrounding UAV operations in real‐world settings has become an urgent and complex challenge. At present, the main methods to minimise the risk of drone decision‐making include utilising pre‐established control rules, expert prior knowledge and regularisation constraints. However, these methodologies require UAVs to meet demanding prerequisites, including the acquisition of extensive decision‐making experience and the establishment of comprehensive rules. Regrettably, these strict requirements often lead to frequent UAV crashes in uncertain environments and subsequent mission failures. In order to tackle these issues, we propose a self‐decision‐making method for quadcopter UAVs based on safe reinforcement learning. Our method utilises a multilevel cascading feature semantic space for reinforcement learning, integrating depth images, greyscale images, semantic segmentation images and object detection results as inputs. This approach aims to facilitate safe autonomous learning. Moreover, we integrate real offline labelled data to enhance the safety policy. Depending on the varying levels of risk encountered during the UAV's decision‐making process, we dynamically select different safety policies. Through this iterative process, the UAV progressively eliminates extreme actions and reverts to the UAV learning policy module. Experimental results indicate that our method not only ensures safe decision‐making for UAVs in uncertain environments but also exhibits superior safety decision‐making efficacy compared to certain baseline methods.

The mass volume of data in the modern business world requires fraud detection to be automated. Hence, some researchers constructed the fraud scenario into graph data and proposed graph-based fraud detection methods. These methods treat the problem of fraud detection as a binary node classification task. However, the differences between the nodes of the same class are ignored. In this paper, we try to distinguish differences in behavior among nodes of the same class to improve the model’s ability to detect deviation, i.e., we make a fine-grained classification of user behavior (called prototypes) and propose an adaptive prototype-based graph neural network (APGNN) for fraud detection. APGNN learns node behavior representations by extracting both neighborhood and global information, supplying preliminary knowledge for the adaptive creation of several prototypes, each representing a distinct behavior pattern. Subsequently, a new loss function is employed to enhance the prototypes’ capacity to capture these behavior patterns and to amplify the feature differences between different prototypes. Nodes are then projected onto these prototypes to derive the final behavior patterns. Extensive experiments on four real-world datasets show that this method can provide better fraud detection as well as a more understandable result.

Graph-based fraud detection has gained significant attention due to its effectiveness in risk management. Recent approaches utilizing graph neural networks highlight the benefits of representing nodes through aggregating homogeneous neighbors. However, in real-world situations, fraudsters disguise their activities by forming connections with benign entities, leading to limitations in the homogeneous aggregation method. Because combining representations from neighborhoods of different categories can dilute the effectiveness of the aggregation. On the other hand, some methods neglect quality of labels and introduce uniform weights for all training nodes, producing challenges for model robustness. To tackle these challenges, we propose a robust model for graph-based fraud detection that leverages adversarial reweighting to improve neighborhood aggregation. In graph-based fraud detection, account nodes can be classified into two types: normal and abnormal. Our model enables dual encoders to assimilate information from both normal and abnormal neighbors, respectively, ensuring resilience against different classes of neighbors. For neighbors without labels, we employ pseudolabels to ascertain the weights for aggregated features across different category spaces. This process still introduces information contamination due to the quality of labeled nodes. To mitigate this issue, we introduce perturbation strategies based on edge structures and node features. These strategies dynamically adjust node weights during training based on the accuracy of model predictions after the introduction of perturbation. This adjustment enhances the contribution of reliable nodes in the loss function, ultimately improving the model’s detection capabilities. Our comprehensive evaluation across four public benchmark datasets-Amazon, Yelp, T-Finance, and T-Social-demonstrates that our proposed method outperforms current state-of-the-art techniques.

Emotion motivates behavior. Investigating the correlation between behavior and emotion, an often overlooked perspective, plays a significant role in uncovering the underlying motives behind behaviors and the intrinsic cause-effects of social events. This article proposes a methodology for mining the correlation between public behavior and emotion using daily news data. Initially, aspect-emotion-reaction (A-E-R) triplets are extracted and generalized, encompassing both explicit and implicit patterns. Then, a knowledge representation model based on hypothetical context (KRHC) with a self-reflection mechanism is proposed to uncover implicit relationships between emotion and behavior through attention mechanisms. By combining rule-based methods for explicit relationships and deep learning for implicit ones, an understanding of emotion-behavior patterns is achieved. In this study, the behaviors are divided into three categories of prosocial, antisocial, and normal behaviors with ten secondary types. Seven categories of emotions are adopted. The proposed deep learning model KRHC is validated on A-E-R datasets and public KINSHIP datasets. The experiment results are concluded; for example, when "fear," "sad," and "surprise" emotions appear, it drives behavior "panic" with most probability. These findings could provide insights for both human-computer interaction and public safety management applications.

In recent years, graph-based fraud detection methods have garnered increasing attention for their superior ability to tackle the issue of camouflage in fraudulent scenarios. However, these methods often rely on a substantial proportion of samples as the training set, disregarding the reality of scarce annotated samples in real-life scenarios. As a theoretical framework within semi-supervised learning, the principle of consistency regularization posits that unlabeled samples should be classified into the same category as their own perturbations. Inspired by this principle, this study incorporates unlabeled samples as an auxiliary during model training, designing a novel barely supervised learning method to address the challenge of limited annotated samples in fraud detection. Specifically, to tackle the issue of camouflage in fraudulent scenarios, we employ disentangled representation learning based on edge information for a small subset of annotated nodes. This approach partitions node features into three distinct components representing different connected edges, providing a foundation for the subsequent augmentation of unlabeled samples. For the unlabeled nodes used in auxiliary training, we apply both strong and weak augmentation and design regularization losses to enhance the detection performance of the model in the context of extremely limited labeled samples. Across five publicly available datasets, the proposed model showcases its superior detection capability over baseline models.

Recent trends in next-item recommendation systems have focused on modeling user intents. Traditional methods often extract users' inherent intents from the most representative items in a session, overlooking “unexpected items” that deviate from the majority in various contextual aspects. These unexpected items, frequently present, can be crucial indicators of a user's inclination towards exploring new options, signaling emerging intents that warrant significant attention. In response, we introduce DbMei, a novel approach that decouples user behaviors and emphasizes the modeling of emerging intents. DbMei distinguishes between two user behavior types: “focused shopping”, which aligns with users' inherent intents, and”wandering shopping”, which aligns with emerging intents. Focused shopping is analyzed using topic modeling and hypergraph learning while wandering shopping is explored through session neighbor retrieval. An exploitation-exploration mechanism is employed to determine the behavioral probability distribution for upcoming items. This integrated modeling of focused and wandering shopping behaviors drives our recommendation process. Extensive empirical studies on two real-world datasets, Amazon-KDD and Beauty, showcase DbMei's superiority over leading methods regarding Recall and MRR metrics. Our code is publicly available at https://github.com/sunlingdan-123/DbMei.

Simulation to reality (sim-to-real) transfer is a promising alternative for training behavioral policies in reinforcement learning (RL). However, many significant differences between the simulator and the real-world environment cause policies to make inconsistent actions in simulation and reality. So policies trained in simulation often perform poorly in real-world. Researchers have explored various methods to bridge this gap, including building highly realistic simulators, implementing automated domain randomization, and employing domain adaptation method. However, highly realistic simulators and automated domain randomization methods rely heavily on extensive real-world data and complex manual processes. Domain adaptation methods require the collection and annotation of high-precision real-world data, leading to low learning efficiency and restricting these methods to specific domains. To address these challenges, this paper proposes transforming the discrepancies in the transfer process into a problem of action sequence similarity. By enhancing the similarity between policy action sequences, we aim to reinforce the consistency of policy actions made in both simulation and reality. For the challenging issue of annotating real-world data, we employ a Generative Adversarial Network (GAN) framework to construct a sim-to-real consistency loss function, thus avoiding reliance on precise real-world data sampling and calibration. To avoid a large amount of real-world data sampling, we introduce Bayesian optimization to accurately and efficiently search for the optimal parameters of the compensation module. Through extensive experiments in multiple sim-to-sim scenarios as well as sim-to-real scenarios, we demonstrate that our method significantly reduces the precision and quantity requirements for real-world data sampling while maintaining high transfer performance.

Fact-based Visual Question Answering (F-VQA) aims to answer questions with observed images and external facts. The existing deep learning-based F-VQA methods still struggle to achieve a fine-grained selection when fusing external knowledge. Furthermore, there is a lack of joint learning between global contextual features and regional features relevant to questions. These limitations weaken the role of external knowledge and hinder the model’s ability to comprehend questions and answers. To address the aforementioned issues, we propose a Hierarchical Knowledge Fusion Network for the F-VQA task. The knowledge fusion comprises the Relation Layer, the Target Layer, and the Reasoning Layer, which aggregate valuable information from questions, images, and facts, respectively. Additionally, we design a Dual Semantic Relations Module to learn object-level contextual relations and pixel-level cross-modal relations simultaneously. Experimental results show that our model achieves state-of-the-art performance on the F-VQA task.

Fraud detection is essential across various domains, including finance and societal contexts. However, the domain of financial fraud detection is plagued by challenges such as class-imbalance, camouflage, and structural fraud. Predominant fraud detection models focus on the former issues, neglecting the nuanced intricacies of structural fraud within financial relations. For instance, in credit card cash-out fraud, fraudulent merchants attempt to blend their illicit transactions by engaging with multiple genuine users, which introduces camouflage. More critically, first-order neighbors of these fraudulent entities are often heterophilic, while their second-order neighbors typically exhibit homophily due to the reliance on intermediary accounts for illegal fund transfers, eventually directing these funds to controlled destination accounts or facilitating money laundering. The role of homophilic and heterophilic relations in multi-hop neighborhoods is paramount for the effective detection of financial fraud. Traditional Graph Neural Networks (GNNs) operate on the homophily assumption, the tendency of nodes to connect with similar others, representing nodes via neighborhood aggregation. They, however, fall short in distinguishing heterophilic interactions, leading to noise incorporation and missing valuable multi-hop neighborhood insights. To address this gap, we propose a novel algorithm named HHSGT (Homophilic and Heterophilic-Aware Sparse Graph Transformer) capable of discerning and emphasizing homophilic interactions while attenuating heterophilic noise within multi-hop neighborhoods. Our approach integrates global attribute information, computes relation scores, allows dynamic score thresholding, and employs a Sparse Graph Transformer to obtain node representation. Extensive experiments on real-world financial fraud, review, and simulated credit card cash-out dataset showcase HHSGT’s superiority over baseline methods.

Jinpeng Li, Hang Yu, Xiangfeng Luo, Qian Liu. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

Summary Event schema refers to the use of a template to depict similar events, and it is a necessary prerequisite for event causality extractions. The induction of event schemas is a difficult task, especially for texts in the open domain, due to the complex and diverse manifestations of events. Previous models considered participants in event mentions are independent or compositional, ignoring the high‐order correlations among participants, which limit their capability of induce event schema. To remedy this, we propose constructing an Event Structure Hypergraph (ESH) to better utilizes the event structural information for event schema induction. In particular, we first extract event mentions from the open‐domain corpus. and then construct an ESH by representing event mentions as a hyperedges. ESH contains high‐order information between participants in event mention. To, learn event mentions representation based on ESH, we propose a weighted attentive hypergraph neural network (WHGNN) to model event high‐order correlations and then integrate node‐category weight matrix into the training of network by improving event representation. By applying jointly cluster algorithm on the event mentions representation, we can induce reliable event schemas. Experimental results on three datasets demonstrate that our approach can induce salient and high‐quality event schemas on open‐domain corpus.

Continuous reinforcement learning carries potential security risks when applied in real-world scenarios, which could have significant societal implications. While its field of application is expanding, the majority of applications still remain confined to virtual environments. If only a single continuous learning method is applied to an unmanned system, it will still forget previously learned experiences, and retraining will be required when it encounters unknown environments. This reduces the learning efficiency of the unmanned system. To address these issues, some scholars have suggested prioritizing the experience playback pool and using transfer learning to apply previously learned strategies to new environments. However, these methods only alleviate the speed at which the unmanned system forgets its experiences and do not fundamentally solve the problem. Additionally, they cannot prevent dangerous actions and falling into local optima. Therefore, we propose a dual decision-making continuous learning method based on simulation to reality (Sim2Real). This method employs a knowledge body to eliminate the local optimal dilemma, and corrects bad strategies in a timely manner to ensure that the unmanned system makes the best decision every time. Our experimental results demonstrate that our method has a 30% higher success rate than other state-of-the-art methods, and the model transfer to real scenes is still highly effective.

In the field of social computing, the task of aspect-based sentiment analysis (ABSA) aims to classify the sentiment polarity of a given aspect in a sentence. The absence of explicit opinion words in the implicit aspect sentiment expressions poses a greater challenge for capturing their sentiment features in the reviews from social media. Many recent efforts use dependency trees or attention mechanisms to model the association between the aspect and other contextual words. However, dependency tree-based methods are inefficient in constructing valuable associations for sentiment classification due to the lack of explicit opinion words. In addition, the use of attention mechanisms to obtain global semantic information easily leads to an undesired focus on irrelevant words that may have sentiments but are not directly related to the specific aspect. In this article, we propose a novel prototype-based demonstration (PD) model for the ABSA task, which contains prototype learning and PD stages. In the prototype learning stage, we employ mask-aware attention to capture the global sentiment feature of aspect and learn sentiment prototypes through contrastive learning. This allows us to acquire comprehensive central semantics of the sentiment polarity that contains the implicit sentiment features. In the PD stage, to provide explicit guidance for the latent knowledge within the T5 model, we utilize prototypes similar to the aspect sentiment as the neural demonstration. Our model outperforms others with a 1.68%/0.28% accuracy gain on the Laptop/Restaurant datasets, especially in the ISE slice, showing improvements of 1.17%/0.26%. These results confirm the superiority of our PD-ABSA in capturing implicit sentiment and improving classification performance. This provides a solution for implicit sentiment classification in social computing.

Due to the overarching similarities of ships, subtle information is imperative for fine-grained ship detection. However, this information is easily lost in adverse weather (e.g., fog, rain, snow, and cloud) or occlusion scenarios. Experts can quickly and accurately recognize fine-grained objects because they have the domain knowledge to help them find the most discriminative information (e.g., edge, structure, texture, and class semantics); thus, they do not need a lot of information to make an identification. Motivated by it, we propose a discriminative information enhancement method with cross-modal domain knowledge (DIE-CDK) for fine-grained ship detection. The core idea behind DIE-CDK is to enhance the discriminative information about fine-grained ships by fusing cross-modal domain knowledge. The introduced cross-modal domain knowledge comprises local and global knowledge: 1) local knowledge is the knowledge of visual shape (e.g., edge contour) which is extracted from the image domain; and 2) global knowledge is the knowledge of the class semantics which is obtained from the common sense domain. In addition, to further study fine-grained ship detection, we introduce a Fine-grained ship dataset (called FgShips). Experiments show that our proposed DIE-CDK method achieves impressive gains in detection performance and outperforms state-of-the-art methods on fine-grained ship and public datasets.

Concept drift is a phenomenon where the distribution of data streams changes over time. When this happens, model predictions become less accurate. Hence, models built in the past need to be re-learned for the current data. Two design questions need to be addressed in designing a strategy to re-learn models: which type of concept drift has occurred, and how to utilize the drift type to improve re-learning performance. Existing drift detection methods are often good at determining when drift has occurred. However, few retrieve information about how the drift came to be present in the stream. Hence, determining the impact of the type of drift on adaptation is difficult. Filling this gap, we designed a framework based on a lazy strategy called Type-Driven Lazy Drift Adaptor (Type-LDA). Type-LDA first retrieves information about both how and when a drift has occurred, then it uses this information to re-learn the new model. To identify the type of drift, a drift type identifier is pre-trained on synthetic data of known drift types. Furthermore, a drift point locator locates the optimal point of drift via a sharing loss. Hence, Type-LDA can select the optimal point, according to the drift type, to re-learn the new model. Experiments validate Type-LDA on both synthetic data and real-world data, and the results show that accurately identifying drift type can improve adaptation accuracy.

Despite significant advances in multi-agent reinforcement learning (MARL) in recent years, learning about cooperative policies for complex tasks remains a challenge. The concept of task allocation provides a valuable tool for designing and comprehending intricate multi-agent systems, as it enables agents with similar roles to collaborate on the same subtasks. However, existing research efforts have not yet provided clear guidelines on how task allocation can better adapt to dynamic environments. This paper introduces a novel framework called Hierarchical Cooperative Task Allocation (HCTA) that is based on long-term behavior analysis. The proposed framework is designed to solve the dynamic task allocation problem in fully cooperative tasks within multi-agent systems. Firstly, we train a task selector based on action chains to enhance the efficiency of task allocation. Secondly, we introduce a hierarchical cooperative mechanism to realize the parallel training of upper-level task selection and lower-level task decision making, so as to accelerate the learning of cooperative strategy. Experimental results show that HTCA significantly improves the learning efficiency of joint policy through dynamic task allocation, and enhances the success rate in complex cooperative tasks in StarCraft II micromanagement benchmark.

Legal Case Recommendation (LCR) is to find out the documents that are most similar to the input case from the judicial point of view. Since the legal documents are long texts and have strong legal attributes, the traditional recommendation method based on text similarity is difficult to accurately understand the legal documents, resulting in poor effect of LCR. To address this problem, we propose Recursive Graph Representation Learning (ReGRL) to hierarchically learn the information in the graph, and obtain a more informative graph representation to accurately understand the case. ReGRL captures nodes, edge, and community information at different levels to integrate information of different granularity. To achieve this, ReGRL performs top-down graph decomposition and bottom-up graph encoding in a recursive form, which allows ReGRL to flexibly control the depth of the learning layers and provide accurate case representations for LCR. Experimental results show that ReGRL can not only generate a good representation, but has a better performance compared to text representation methods for LCR. In addition, we also performed different experiments to analyze the principle of ReGRL and verify the effectiveness of recursive procedures.

Multimodal emotion recognition comprehensively identifies the emotion contained in multimodal data by bridging the gaps between heterogeneous dataset. In recent years, multimodal emotion recognition methods have gained significant attention and been shown to surpass single-modal approaches. Most of the existing multimodal emotion analysis methods simply combine different modalities to improve the recognition capability of consistent emotion expressions across multimodal data. However, it remains challenging to recognize the right emotion when multiple modalities' contents carry inconsistent or even contradictory emotions. To solve this problem, we propose a novel image-text emotion recognition model named Multimodal Cross-Attention Bayesian Network(MCABN). The entire network exploits Bayesian theory to learn the distribution of its weight parameters, making optimization directions and results of parameters interpretable. What's more, the model leverages the consistency and complementarity between visual content and textual description to arrive at accurate decisions. Specifically, for each modality, multiple explainable features (color, texture, and shape feature in the image, while adjective, adverb, verb, noun, and negative feature in the text) and one unexplainable feature are fused as its feature representation to reinforce emotion-related features. Then two single-modal attention modules(Visual Attention Module and Textual Attention Module) capture the most discriminative features in a single image and text; Two cross-modal attention modules(Image-guided Text Attention Module and Text-guided Image Attention Module) extract the complementary and dominant features between two modalities by interactive learning. Finally, the outputs of four attention modules are integrated through intermediate fusion to predict the final emotion. The experimental results on the NVTD and MVSA-Multiple dataset indicate that the proposed MCABN outperforms state-of-the-art baselines by substantial margins.

Event Extraction (EE) is an important part of Information Extraction (IE), whose primary mission is to automatically extract specific events and event-related arguments from the original text. This is helpful to classify, extract and reconstruct massive content. At present, most EE methods aim at single event extraction based on neural networks. However, some sentences contain multiple events, and a single argument may play different roles in different events, which is easy to cause the problem of argument role overlap. Furthermore, the latest proposed prompt learning requires too much template annotation cost. Therefore, this paper proposes a new EE method based on the self-attention mechanism, which obtains global sentence semantics by capturing the Dynamic Prompt Information (DPI) without additional manual annotation. In addition, we integrate the local sentence semantics of Multi-dimensional Event Feature (MEF) information with DPI to learn the internal correlation information between different events and different entities, so as to realize the extraction of argument multiple roles. Experiments on the widely used Automatic Content Extraction (ACE) English corpora show that our method achieves a substantial improvement over baselines. Additionally, further analysis of the extraction procedure is presented to illustrate the effectiveness of the proposed DPI and MEF.

LiDAR-based 3D object detection provides the necessary high-precision environmental sensing information for the safe navigation of smart ships.However, relying on viewpoint projections, voxelized point clouds, or using inefficient point sampling methods, current LiDAR 3D object detection methods treat all objects uniformly and quantitatively while ignoring the specificity of sparse objects in the scene, which leaves less useful information about sparse objects.In this paper, we propose an end-to-end two-stage architecture, Object-Level Contrast Learning 3D Object Detection network (OCL), for better construction of sparse object features and improving the ability of model to detect sparse objects.In the first stage, the Contrast Learning based Sparse Object Feature Enhancement training strategy is proposed to decrease the feature discrepancy between sparse and regular objects in object-level.In the second stage, we use the Point-level Feature Multiple Aggregation Strategy to aggregate finer point-level features of sparse objects.Extensive experiments show that OCL achieves excellent performance on both Ship dataset and KITTI dataset.Furthermore, our work proposes a promising new idea for applying contrast learning to 3D object detection.

Few-shot object detection (FSOD), which involves training the detector with few annotated data to detect novel objects, has aroused a wide range of research interests.However, the performance of FSOD is still limited by insufficient data.Existing works usually adopt fine-tuning paradigm, which first uses rich base classes for pre-training and then uses them to carve the novel class feature space.In the fine-tuning phase, the balance space learned by the pre-trained model will be broken leading to an intersection between the feature space of novel and base classes, which makes it difficult to distinguish the difference between them.Contrastive learning has been shown to learn a balanced feature space and enhance the discriminability of the learned features.Here, we present Few-Shot object detection via Instance-wise and Prototype Contrastive Learning (FS-IPCL), which introduces contrastive learning to learn a balanced feature space.FS-IPCL uses instance-wise and prototype contrastive loss during feature learning to enhance the intra-class compactness and inter-class separability of samples.In this way, the base and novel classes can be evenly distributed in the feature space, improving the class boundary to alleviate the confusion problem of the novel classes.Extensive experimental results on the PASCAL VOC and MS-COCO datasets demonstrate the effectiveness of the proposed method and achieve state-of-theart performance.

Few-shot link prediction is an important recent research direction in the field of knowledge graph.To learn better entity and relation representations, previous works utilize the neighborhood information of entities but they ignore the associated relations.More than this, the link invovles few-shot entities cannot be well linked by these methods.Therefore, this paper proposes a novel method that dynamically aggregates local neighbourhood information from entities and relations and introduces the information bottleneck principle to filter irrelevant features to better capture the feature of few-shot entities.Experimental results on three benchmark datasets demonstrate that our model outperforms other models and that our approach can effectively improve the model performance on those entities in the tail of the long-tailed distribution.

Continuous reinforcement learning carries potential security risks when applied in real-world scenarios, which could have significant societal implications.While its field of application is expanding, the majority of applications still remain confined to virtual environments.if only a single continuous learning method is applied to an unmanned system, it will still forget previously learned experiences, and retraining will be required when it encounters unknown environments.This reduces the learning efficiency of the unmanned system.To address these issues, some scholars have suggested prioritizing the experience playback pool and using transfer learning to apply previously learned strategies to new environments.However, these methods only alleviate the speed at which the unmanned system forgets its experiences and do not fundamentally solve the problem.Additionally, they cannot prevent dangerous actions and falling into local optima.Therefore, we propose a dual decision-making continuous learning method based on Simulation to Reality (Sim2Real).This method employs a knowledge body to eliminate the local optimal dilemma, and corrects bad strategies in a timely manner to ensure that the unmanned system makes the best decision every time.Our experimental results demonstrate that our method has a 30% higher success rate than other state-of-the-art methods, and the model transfer to real scenes is still highly effective.

Answer selection aims to identify the most relevant answers to a given question from a set of candidates. It is the fundamental component of intelligent question answering system. To improve performance, it gradually becomes an effective strategy to integrate external structured knowledge bases (KBs) into the answer selection model. Due to expensive cost of construction and maintenance of such KBs, these models are suffering from domain barriers and information incompleteness. In this paper, we propose a two-stage extraction–comprehension answer selection model, which can extract ad-hoc knowledge from unstructured web texts to enhance the performance of answer selection. For the extraction, two types of snippets are extracted from unstructured web pages and utilized as the source of ad-hoc knowledge. For the comprehension, a selective attention mechanism is employed to extract and integrate ad-hoc knowledge from multiple text snippets obtained in the first stage, which can enrich the representation of question–answer pairs and more accurately identify the correct answers. By incorporating ad-hoc knowledge extracted from both types of snippets, the proposed model achieves state-of-the-art results on two public available benchmark datasets. In particular, on WikiQA, in terms of the two evaluation metrics (mean average precision and mean reciprocal rank), it achieves 9.9[Formula: see text] and 8.4[Formula: see text] higher than the previous non-pretraining-based models, and 3.4[Formula: see text] and 3.2[Formula: see text] higher than the pretraining-based models.

The real-time deployment of bidirectional encoder representations from transformers (BERT) is limited by its slow inference caused by its large number of parameters. Recently, multi-exit architecture has garnered scholarly attention for its ability to achieve a trade-off between performance and efficiency. However, its early exits suffer from a considerable performance reduction compared to the final classifier. To accelerate inference with minimal compensation of performance, we propose a novel training paradigm for multi-exit BERT performing at two levels: training samples and intermediate features. Specifically, for the training samples level, we leverage curriculum learning to guide the training process and improve the generalization capacity of the model. For the intermediate features level, we employ layer-wise distillation learning from shallow to deep layers to resolve the performance deterioration of early exits. The experimental results obtained on the benchmark datasets of textual entailment and answer selection demonstrate that the proposed training paradigm is effective and achieves state-of-the-art results. Furthermore, the layer-wise distillation can completely replace vanilla distillation and deliver superior performance on text entailment datasets.

Abstract In recent years, the extraction of overlapping relations has received great attention in the field of natural language processing (NLP). However, most existing approaches treat relational triples in sentences as isolated, without considering the rich semantic correlations implied in the relational hierarchy. Extracting these overlapping relational triples is challenging, given the overlapping types are various and relatively complex. In addition, these approaches do not highlight the semantic information in the sentence from coarse-grained to fine-grained. In this paper, we propose an end-to-end neural framework based on a decomposition model that incorporates multi-granularity relational features for the extraction of overlapping triples. Our approach employs an attention mechanism that combines relational hierarchy information with multiple granularities and pretrained textual representations, where the relational hierarchies are constructed manually or obtained by unsupervised clustering. We found that the different hierarchy construction strategies have little effect on the final extraction results. Experimental results on two public datasets, NYT and WebNLG, show that our mode substantially outperforms the baseline system in extracting overlapping relational triples, especially for long-tailed relations.

Graphs have been widely adopted to accomplish fraud detection tasks because of their inherently favorable structure to capture the intricate features in many complicated scenarios, especially in some modern e-commerce situations that have various relation attributes like transactions. These works tend to utilize the direct aggregate information about neighbor nodes of the target node or the aggregation of the neighbor information after the conditional filter and mostly use local information but ignore global information. However, in some cases, local abnormal points are detected in the global view, so global information is very important for fraud detection. In this article, we propose a local and global aware memory-based graph neural network for fraud detection (LGM-GNN). It first obtains the preliminary node embedding through relation-aware embedding and then interactively aggregates the local and global memory network to fuse and utilize the local and global information. Finally, the node embeddings of different levels are aggregated through the hierarchical information aggregator. Extensive experiments show our proposed LGM-GNN outperforms other SOAT methods on two real-world fraud detection datasets.

Robots trained by Imitation Learning(IL) are used in many tasks(e.g., autonomous vehicle manipulation). Generative Adversarial Imitation Learning (GAIL) assumes that the demonstration set used for training is of high quality. However, such demonstrations are difficult and expensive to obtain. GAIL-related methods fail to learn effective strategies if non-high quality demonstrations are used because the performance of agents trained by this method is limited by the demonstrator's operations. Our idea is to enable the agent to learn strategy with better performance than the demonstrator from a suboptimal demonstration set, which contains non-high quality demonstrations that are easier to obtain. Inspired by this, we propose the Trajectory-Ranked Adversarial Imitation Learning (TRAIL) method. First, for demonstration set processing, we introduce a ranking process and define the concept of Performance Relative Advantage of suboptimal demonstrations to specify the ranking order. Second, for model training, we reconstruct the objective function of GAIL and use an experience replay buffer, enabling the agent to learn implicit features and ranking information from the ranked suboptimal demonstration set and possess the ability to outperform the demonstrator. Experiments show that in Mujoco's tasks, our method can learn from a suboptimal demonstration set and can achieve better performance than baseline methods.

Multi-agent reinforcement learning (MARL) has been applied to many multi-agent team tasks, such as multi-robot swarm control. Distributional Value Function Factorization (DFAC) follows the Distributional-Individual-Global-Max (DIGM) principle, which forces the individual's optimal action to be in accordance with the optimal joint action at all times. However, this principle leads to grade inflation, which limits agents to exploring better strategies than before. We focus on this issue and propose a novel MARL method named dueling mixing distributional Q-learning with fully parameters (FDMIX). Firstly, a parametric individual value network generates an individual distribution function and a utility value function, while the fractions are obtained through a fraction proposal network. Secondly, the conversion mixing network obeys a new advantage-based DIGM principle to generate a joint distribution action value based on the global state. Finally, we incorporate an N-step return-based loss function to achieve stable and efficient training. Our extensive tests on the multiple-particle environment and StarCraft II show that our method performs better than state-of-the-art algorithms noticeably.

Extracting knowledge from closely-related domains to solve a new problem has become an advanced methodology in machine learning, and is called transfer learning. Conspicuous among existing regression methods are those built around Gaussian process (GP) because they consider variance in the predicted values, although they also require that the similarity between the source domain and the target domain needs to fall within a certain range. To overcome this limitation in GP methods and improve transfer learning performance, this study proposes an adaptive Gaussian process (AGP) for regression. The AGP method broadens the range of acceptable similarity in current GP methods by developing a new transfer kernel. The results of experiments with transfer regression problems on both synthetic and real-world datasets indicate that this AGP method signiticantly improves prediction accuracy.

Abstract Causal event extraction (CEE) aims to identify and extract cause‐effect event pairs from texts, which is a fundamental task in natural language processing. Recent research treat CEE as a sequence labeling problem. However, the linguistic complexity and ambiguity of textual description results in the low accuracy of extractors. To address the above issues, considering the prior knowledge like the causal network constructed based on the causal indicators, which can represent information transition between cause and effect, may helpful for CEE. In this article, we propose causality‐associated graph neural network to incorporate in‐domain knowledge by taking important causal words into account. External causal knowledge is modeled as causal associated graph (CAG). Then we use graph neural networks (GNN) to capture the complex relationship of intraevent mentions and interevent causality in a sentence based on the relationship obtained from CAG. Finally, sentence sequence and prior causal knowledge of GNN embedding are fed into multiscaled convolution and bidirectional long short‐term memory networks. Experimental results on two datasets show that our method outperforms the state‐of‐the‐art baseline.

Abstract Unmanned surface vessels (USVs) have been fully used in the civilian and military fields in recent years, which dramatically expands protective capability and detection range. However, the marine environment's complexity and variability make that verification of various advanced USVs control algorithms face high costs and high risks. In this article, we present USVs‐Sim, a novel high‐fidelity general simulation platform for USVs autonomous navigation data generation and control strategy testing. USVs‐Sim is a collection of high‐level extensible modules that allows the rapid development and testing of USVs configurations and facilitates the construction of complex ocean scenarios. USVs‐Sim supports the steering or thrusting limits of USVs, as well as unique dynamics profiles. The platform can specify specific USVs sensor systems and change the time of day and weather conditions to generate robust data. USVs‐Sim facilitates training of deep‐learning algorithms by enabling data export from USVs sensors, including vision data, lidar, relative positions of ocean targets. Therefore, USVs‐Sim allows for the rapid prototyping, development, and testing of USVs autonomous control algorithms in a complex marine environment. In this article, we detail the general simulation platform and testing several representative USVs intelligent control algorithms on the platform.

Abstract The inherent disorder and irregularity of 3D point clouds pose great challenges to classification and segmentation tasks. To tackle these problems, we propose a geometric relation based point clouds classification and segmentation network. Specifically, we design two novel modules named geometric relation based convolution (GRC) and relational attention interpolation (RAI) to infer the local relations of point clouds. In GRC module, the convolutional weights and local features are both reasoned from predefined local geometric relations between the central point of each local point clouds and its neighboring points. The global shape awareness is obtained by stacking convolutional layers of the GRC. In RAI, a relational attention interpolation approach is proposed for the segmentation task. The attentional weights of different neighboring points are inferred from local relations of geometry and features, which is capable of guiding RAI to pay more attention to the relevant points and be sensitive to the boundaries of segments. Experimental results show that the proposed method makes full use of the geometric relations between local points, and presents good performance on both classification and segmentation tasks.

Abstract Link prediction plays an important role in constructing knowledge graph. Recently, graph representation learning models yield state‐of‐the‐art results. However, existing models concentrate merely on triples or graph structures and mostly ignore textual descriptions, resulting in incomplete or partial information. In this paper, we propose a novel graph representation learning model to address this challenge, namely multi‐source heterogeneous information fusion with adaptive importance sampling. Our model leverages multiple sources, such triple, graph structure and textual description, and generate rich‐attribute embeddings for entities, encapsulating relations simultaneously. We also propose an adaptive importance sampling algorithm to boost aggregation of useful features from local neighbours. Additionally, we also boost node aggregation of useful features from local neighbours by adaptive importance sampling algorithm in our model. Experimental results on two benchmark datasets show that our proposed model significantly outperforms state‐of‐the‐art methods.

Abstract Event causality extraction is a challenging task in natural language processing (NLP), which plays an important role in event prediction, scene generation, question answering and textual entailment. Most existing methods focus on extracting single‐scale (such as phrase) event causality, while fails to extract multi‐scale (such as word, phrase, sentence) event causality. To fill the gap, we propose multi‐scale event causality extraction via simultaneous knowledge‐attention and convolutional neural network (KA‐CNN). First, knowledge‐attention takes N‐gram embedding as input and takes semantic features, fused with prior knowledge through causal associative link network (CALN), as output. Second, multi‐scale CNN is designed with word embedding as input and semantic feature of corpus as output. Third, bidirectional long short‐term memory with conditional random field (BiLSTM + CRF) is conducted after concatenation of features from knowledge‐attention and multi‐scale CNN. Finally, we compare our results with other baselines. The experimental results show that our proposed method shows promising result in extracting multi‐scale event causality.

The understanding of people’s inter-regional mobility behaviors, such as predicting the next activity region (AR) or uncovering the intentions for regional mobility, is of great value to public administration or business interests. While there are numerous studies on human mobility, these studies are mainly from a statistical view or study movement behaviors within a region. The work on individual-level inter-regional mobility behavior is limited. To this end, in this article, we propose a dynamic region-relation-aware graph neural network (DRRGNN) for exploring individual mobility behaviors over ARs. Specifically, we aim at developing models that can answer three questions: (1) Which regions are the ARs? (2) Which region will be the next AR, and (3) Why do people make this regional mobility? To achieve these tasks, we first propose a method to find out people’s ARs. Then, the designed model integrates a dynamic graph convolution network (DGCN) and a recurrent neural network (RNN) to depict the evolution of relations between ARs and mine the regional mobility patterns. In the learning process, the model further considers peoples’ profiles and visited point-of-interest (POIs). Finally, extensive experiments on two real-world datasets show that the proposed model can significantly improve accuracy for both the next AR prediction and mobility intention prediction.

Causal Event Extraction (CEE) is a joint extraction task of events and causality, which can help text understanding, event prediction and so on. Recent research has achieved state-of-the-art performance in various Natural Language Processing (NLP) tasks by combining pre-trained models with neural networks. However, ambiguity of event description and long-distance dependence of event causality result in the low accuracy of extractors. In this paper, we propose a model to incorporate in-domain knowledge by taking frequent expression of event causality into account, and use iterated dilated convolutions to expand the perception field of event causality. External causal knowledge is modeled as frequent n-grams with different length, which is used as convolution filters during kernel initialization, enhancing the ability of model to capture the boundary of event description. To obtain long-distance dependence of event causality, we use iterated dilated convolutions to aggregate context from the entire sentence. Experimental results show that our method significantly outperform the baselines with faster convergence speed.

Due to the success of Bidirectional Encoder Representations from Transformers (BERT) in natural language process (NLP), the multi-head attention transformer has been more and more prevalent in computer-vision researches (CV). However, it still remains a challenge for researchers to put forward complex tasks such as vision detection and semantic segmentation. Although multiple Transformer-Based architectures like DETR and ViT-FRCNN have been proposed to complete object detection task, they inevitably decreases discrimination accuracy and brings down computational efficiency caused by the enormous learning parameters and heavy computational complexity incurred by the traditional self-attention operation. In order to alleviate these issues, we present a novel object detection architecture, named Convolutional vision Transformer-Based Attentive Single Shot MultiBox Detector (CvT-ASSD), that built on the top of Convolutional vision Transormer (CvT) with the efficient Attentive Single Shot MultiBox Detector (ASSD). We provide comprehensive empirical evidence showing that our model CvT-ASSD can leads to good system efficiency and performance while being pretrained on large-scale detection datasets such as PASCAL VOC and MS COCO. Code has been released on public github repository at https://github.com/albert-jin/CvT-ASSD.

In a mixed cooperative-competitive environment, each agent needs to learn heterogeneous strategies. The complex game relationship between heterogeneous agents causes difficulties for strategy learning. In this paper, we propose a graph-based multi-agent reinforcement learning method to simplify the learning process, named hierarchical heterogeneous graph multi-agent actor-critic (H2G-MAAC). The method first uses a hierarchical heterogeneous graph to model the hierarchical relationships among multiple heterogeneous agents. Then, it conducts representation learning for multiple agents with hierarchical graph attention networks. Finally, it learns heterogeneous strategies with multi-agent actor-critic. We conduct experiments in Predator-Prey games. The results indicate that the proposed method can simplify the learning process and outperforms existing state-of-the-art methods.

Summary Event evolutionary graph (EEG) reflects sequential and causal relations between events, which is of great value for event prediction. However, lacking event context in the EEG raises the problems of direction uncertainty and low accuracy when making predictions. In this article, we propose a conditional event evolutionary graph (CEEG) to deal with these problems. CEEG extends EEG with an additional four types of event context, including state, cause, sub‐type, and object. We first extract event context by matching the input with self‐adaptive semantic templates and generalize the context for each event. To identify the evolution direction, we treat it as a binary classification problem and calculate the event transition probability for each direction given the generalized context. Experimental results show that CEEG has a strong ability to generate better event evolutionary paths compared with NAR, EEM, and other non‐context‐based methods.

Abstract Autonomous obstacle avoidance control of unmanned surface vessels (USVs) in complex marine environments is always fundamental for its scientific search and detection. Traditional methods usually model USV motion and environments in a mathematical way that needs perceptual information. Unfortunately, it is difficult to provide sufficient perceptual information due to complex marine environments, resulting in inaccurate modeling. Reinforcement learning has recently enjoyed increasing popularity in the problem of obstacle avoidance since it can settle problems by partially observable environment information. However, the autonomous USV obstacle avoidance using reinforcement learning is still facing the challenge of designing appropriate reward functions under complex marine environments. To address these issues, we propose a prior knowledge‐based USV reinforcement learning obstacle avoidance algorithm. In this algorithm, an actor‐critic network is used as the main architecture of the algorithm, and prior knowledge‐based reward shaping used to design relevant reward function for USV obstacle avoidance. A standard USV based on a visual sensor is designed, and the state input of the algorithm is through USV's front vision sensor. We conducted simulation experiments and results prove that our algorithm can effectively converge, and USV achieves high velocity and low collision rate in the complex marine environment.

Learning adversarial policy aims to learn behavioural strategies for agents with different goals, is one of the most significant tasks in multi-agent systems. Multi-agent reinforcement learning (MARL), as a state-of-the-art learning-based model, employs centralised or decentralised control methods to learn behavioural strategies by interacting with environments. It suffers from instability and slowness in the training process. Considering that parallel simulation or computation is an effective way to improve training performance, we propose a novel MARL method called Multiple scenes multi-agent proximal Policy Optimisation (MPO) in this paper. In MPO, we first simulate multiple parallel scenes in the training environment. Multiple policies control different agents in the same scene, and each policy also controls several identical agents from multiple scenes. Then, we expand proximal policy optimisation (PPO) with an improved actor-critic network, ensuring the stability of training in multi-agent tasks. The actor network only uses local information for decision making, and the critic network uses global information for training. Finally, effective training trajectories are computed with two criteria from multiple parallel scenes rather than single to accelerate the learning process. We evaluate our approach in two simulated 3D environments, one of which is Unity's official open-source soccer game, and the other is unmanned surface vehicles (USVs) built by Unity. Experiments demonstrate that MPO converges more stable and faster than benchmark methods in model training, and demonstrates excellent adversarial policy compared with benchmark models.

Abstract Given a question and its answer candidates (named QA corpus), answer selection is the task of identifying the most relevant answers to the question. Answer selection is widely used in question answering, web search, and so on. Current deep neural network models primarily utilize local features extracted from input question‐answer pairs (QA pairs). However, the global features contained in QA corpora are under‐utilized, and we argue that these global features substantially contribute to the answer selection task. To verify this point of view, we propose a novel model that combines local and global features for answer selection. In our model, two different global feature extractors are employed to extract statistical global features and deep global features from a QA corpus, respectively. Furthermore, we investigate the integration of these global features with local features in various experimental settings: statistical global features, deep global features, and a combination of statistical and deep global features. Our experimental results show that the global features are effective for answer selection. Our model obtains new state‐of‐the‐art results on two public answer selection datasets and performs especially well on YahooCQA, where it achieves 9.2 and 6% higher precision@1 (P@1) and mean reciprocal rank (MRR) scores than previously published models.

Emotion analysis has been attracting researchers' attention. Most previous works in the artificial-intelligence field focus on recognizing emotion rather than mining the reason why emotions are not or wrongly recognized. The correlation among emotions contributes to the failure of emotion recognition. In this article, we try to fill the gap between emotion recognition and emotion correlation mining through natural language text from Web news. The correlation among emotions, expressed as the confusion and evolution of emotion, is primarily caused by human emotion cognitive bias. To mine emotion correlation from emotion recognition through text, three kinds of features and two deep neural-network models are presented. The emotion confusion law is extracted through an orthogonal basis. The emotion evolution law is evaluated from three perspectives: one-step shift, limited-step shifts, and shortest path transfer. The method is validated using three datasets: 1) the titles; 2) the bodies; and 3) the comments of news articles, covering both objective and subjective texts in varying lengths (long and short). The experimental results show that in subjective comments, emotions are easily mistaken as anger. Comments tend to arouse emotion circulations of love-anger and sadness-anger. In objective news, it is easy to recognize text emotion as love and cause fear-joy circulation. These findings could provide insights for applications regarding affective interaction, such as network public sentiment, social media communication, and human-computer interaction.

Multi-scale object detection involves classification and regression assignments of objects with variable scales from an image. How to extract discriminative features is a key point for multi-scale object detection. Recent detectors simply fuse pyramidal features extracted from ConvNets, which does not take full advantage of useful features and drop out redundant features. To address this problem, we propose Context-Aware Hierarchical Feature Attention Network (CHFANet) to focus on effective multi-scale feature extraction for object detection. Based on single shot multibox detector (SSD) framework, the CHFANet consists of two components: the context-aware feature extraction (CFE) module to capture rich multi-scale context features and the hierarchical feature fusion (HFF) module followed with the channel-wise attention model to generate deeply fused attentive features. On the Pascal VOC benchmark, our CHFANet can achieve 82.6% mAP. Extensive experiments demonstrate that the CHFANet outperforms a lot of state-of-the-art object detectors in accuracy without any bells and whistles.

Cooperation among agents with partial observation is an important task in multi-agent reinforcement learning (MARL), aiming to maximize a common reward. Most existing cooperative MARL approaches focus on building different model frameworks, such as centralized, decentralized, and centralized training with decentralized execution. These methods employ partial observation of agents as input directly, but rarely consider the local relationship between agents. The local relationship can help agents integrate observation information among different agents in a local range, and then adopt a more effective cooperation policy. In this paper, we propose a MARL method based on spatial relationship called hierarchical relation graph soft actor-critic (HRG-SAC). The method first uses a hierarchical relation graph generation module to represent the spatial relationship between agents in local space. Second, it integrates feature information of the relation graph through the graph convolution network (GCN). Finally, the soft actor-critic (SAC) is used to optimize agents' actions in training for compliance control. We conduct experiments on the Food Collector task and compare HRG-SAC with three baseline methods. The results demonstrate that the hierarchical relation graph can significantly improve MARL performance in the cooperative task.

Knowledge Graph Completion (KGC) aims at completing the missing information in Knowledge Graphs (KGs) that provide a structured representation of knowledge. Most existing methods focus on completing existing entities and relationships that have not been discovered, while neglecting the importance of discovering new knowledge outsides KGs, especially the relationships, to the KGC. In this paper, we work further and try to detect relationship between two entities no matter they are inside or outside KGs. To complete this task, we propose a novel and unified model named Structured Attention Graph Neural Network (SAGNN) which can mine the new relationships outside KGs easily without relying on much external resource. By mining the latent association features from structural information such as in-degree, out-degree and co-occurrence frequency, the model enables specifying different weights to different nodes in a neighborhood even if the enter node lacks embedding representation, and generates the embedding with high-level semantic features eventually. Experiments on multiple datasets show that SAGNN performs well on the open-world relationship prediction task and even outperforms existing KGC models on the triple classification task in which entities outside KGs are involved.

Experience memory, which stores and replays experience from the past for off-policy reinforcement learning (RL) agents, has become an essential component of cutting-edge algorithms in recent years. In most previous work, experience memory emphasizes recently acquired experiences while neglects long-ago ones, which causes high variance and catastrophic forgetting in continual learning. To solve the problem, we introduce the Staged Experience Mechanism(SEM) - a novel management mechanism of experience memory. SEM adaptively regulates the proportion of experiences based on the current learning stages, enabling agents not only to learn from new experiences but also from very old ones given limited memory capacity. In the experiment, SEM is coupled with two state-of-the-art off-policy algorithms compared with classic experience replay mechanisms on the suite of MuJoCo tasks. The result shows that RL with SEM is more stable and less variable than its counterparts in most tested continuous control environments, which reduces the training variance by 5% ∼ 104% while ensuring optimal scoring performance.

Style transfer aims to generate images by combining the style of one image and the content of another. Though valuable efforts have been made in generating high-quality style transferred images, the resulting images are far from the distribution of real images. This greatly limits the application of style transfer such as improving the diversity of training set in computer vision task. We find that the reason of style transfer failing to generate realistic images is lack of reference targets and neglect of preserving data distribution. To solve the problem, we propose a Style-transfer Generative Adversarial Network with a weight-sharing strategy to make the stylized images be resemblance to the real images. The experimental results demonstrate that the proposed method can generate images with satisfying style transfers and high visual quality. Moreover, we apply our stylized images to augment the training set of object detection task, and improve the average precision faithfully. We believe that our method can enhance the performance of style transfer on computer vision tasks.

Summary Taxonomic relations play an important role in various Natural Language Processing (NLP) tasks (eg, information extraction, question answering and knowledge inference). Existing approaches on embedding‐based taxonomic relation learning mainly rely on the word embeddings trained using co‐occurrence‐based similarity learning. However, the performance of these approaches is not quite satisfactory due to the lack of sufficient taxonomic semantic knowledge within word embeddings. To solve this problem, we propose an improved embedding‐based approach to learn taxonomic relations via incorporating relation descriptions into word embeddings. First, to capture additional taxonomic semantic knowledge, we train special word embeddings using not only co‐occurrence information of words but also relation descriptions (eg, taxonomic seed relations and their contextual triples). Then, using the trained word embeddings as features, we employ two learning models to identify and predict taxonomic relations, namely, offset‐based classification model and offset‐based similarity model. Experimental results on four real‐world domain datasets demonstrate that our proposed approach can capture additional taxonomic semantic knowledge and reduce dependence on the training dataset, outperforming the state‐of‐the‐art compared approaches on the taxonomic relation learning task.

Summary Detecting different emotions is a key step toward intelligent recommendation and affective computing, which plays an increasingly important role in guiding decisions. There have been a large number of previous work of machine learning in this topic. However, most existing methods neglect the help of long‐term accumulated and precious interdisciplinary knowledge, which cover both computer science and psychology. In this work, we distill part of the psychological knowledge as following: (i) The emotions with the same polarity co‐occur frequently, while emotions with contrary polarity co‐exist hardly; (ii) The relations between emotions are variable and complicated; (iii) The number of emotions that can be aroused in the same moment is limited. In this paper, we propose two novel modules to execute above prior knowledge, by adding external constraints to our proposed model Double Attention Convolutional Neural Network. In the first module, the emotion relationships are captured by an automatically extracted emotion template. In the second module, uncertainty punishment is applied to limit the too frequent emotions. Our proposed model, which employs interdisciplinary knowledge, performs potentially and get state‐of‐the‐art result when compared with models without above constraints.

In different contexts, one abstract concept (e.g., fruit) may be mapped into different concrete instance sets, which is called abstract concept instantiation. It has been widely applied in many applications, such as web search, intelligent recommendation, etc. However, in most abstract concept instantiation models have the following problems: (1) the neglect of incorrect label and label incompleteness in the category structure on which instance selection relies; (2) the subjective design of instance profile for calculating the relevance between instance and contextual constraint. The above problems lead to false prediction in terms of abstract concept instantiation. To tackle these problems, we proposed a novel model to instantiate the abstract concept. Firstly, to alleviate the incorrect label and remedy label incompleteness in the category structure, an improved random-walk algorithm is proposed, called InstanceRank, which not only utilize the category information, but it also exploits the association information to infer the right instances of an abstract concept. Secondly, for better measuring the relevance between instances and contextual constraint, we learn the proper instance profile from different granularity ones. They are designed based on the surrounding text of the instance. Finally, noise reduction and instance filtering are introduced to further enhance the model performance. Experiments on Chinese food abstract concept set show that the proposed model can effectively reduce false positive and false negative of instantiation results.

Relation classification is one of the most fundamental upstream tasks in natural language processing and information extraction. State-of-the-art approaches make use of various deep neural networks (DNNs) to extract higher-level features directly. They can easily access to accurate classification results by taking advantage of both local entity features and global sentential features. Recent works on relation classification devote efforts to modify these neural networks, but less attention has been paid to the feature design concerning syntax. However, from a linguistic perspective, syntactic features are essential for relation classification. In this article, we present a novel linguistically motivated approach that enhances relation classification by imposing additional syntactic constraints. We investigate to leverage syntactic skeletons along with the sentential contexts to identify hidden relation types. The syntactic skeletons are extracted under the guidance of prior syntax knowledge. During extraction, the input sentences are recursively decomposed into syntactically shorter and simpler chunks. Experimental results on the SemEval-2010 Task 8 benchmark show that incorporating syntactic skeletons into current DNN models enhances the task of relation classification. Our systems significantly surpass two strong baseline systems. One of the substantial advantages of our proposal is that this framework is extensible for most current DNN models.

Crowdsourcing is a process of acquisition, integration, and analysis of big and heterogeneous data generated by a diversity of sources in urban spaces, such as sensors, devices, vehicles, buildings, and human. Especially, nowadays, no countries, no communities, and no person are immune to urban emergency events. Detection about urban emergency events, e.g., fires, storms, traffic jams is of great importance to protect the security of humans. Recently, social media feeds are rapidly emerging as a novel platform for providing and dissemination of information that is often geographic. The content from social media usually includes references to urban emergency events occurring at, or affecting specific locations. In this paper, in order to detect and describe the real time urban emergency event, the 5W (What, Where, When, Who, and Why) model is proposed. Firstly, users of social media are set as the target of crowd sourcing. Secondly, the spatial and temporal information from the social media are extracted to detect the real time event. Thirdly, a GIS based annotation of the detected urban emergency event is shown. The proposed method is evaluated with extensive case studies based on real urban emergency events. The results show the accuracy and efficiency of the proposed method.

State-of-the-art methods for relation extraction consider the sentential context by modeling the entire sentence. However, syntactic indicators, certain phrases or words like prepositions that are more informative than other words and may be beneficial for identifying semantic relations. Other approaches using fixed text triggers capture such information but ignore the lexical diversity. To leverage both syntactic indicators and sentential contexts, we propose an indicator-aware approach for relation extraction. Firstly, we extract syntactic indicators under the guidance of syntactic knowledge. Then we construct a neural network to incorporate both syntactic indicators and the entire sentences into better relation representations. By this way, the proposed model alleviates the impact of noisy information from entire sentences and breaks the limit of text triggers. Experiments on the SemEval-2010 Task 8 benchmark dataset show that our model significantly outperforms the state-of-the-art methods.

Instability and slowness are two main problems in deep reinforcement learning. Even if proximal policy optimization (PPO) is the state of the art, it still suffers from these two problems. We introduce an improved algorithm based on proximal policy optimization, mixed distributed proximal policy optimization (MDPPO), and show that it can accelerate and stabilize the training process. In our algorithm, multiple different policies train simultaneously and each of them controls several identical agents that interact with environments. Actions are sampled by each policy separately as usual, but the trajectories for the training process are collected from all agents, instead of only one policy. We find that if we choose some auxiliary trajectories elaborately to train policies, the algorithm will be more stable and quicker to converge especially in the environments with sparse rewards.

Distantly supervised relation extraction leverages knowledge bases to label training data automatically. However, distant supervision may introduce incorrect labels, which harm the performance. Many efforts have been devoted to tackling this problem, but most of them treat relation extraction as a simple classification task. As a result, they ignore useful information that comes from related tasks, i.e., dependency parsing and entity type classification. In this paper, we first propose a novel Multi-Task learning framework for Relation Extraction (MTRE). We employ dependency parsing and entity type classification as auxiliary tasks and relation extraction as the target task. We learn these tasks simultaneously from training instances to take advantage of inductive transfer between auxiliary tasks and the target task. Then we construct a hierarchical neural network, which incorporates dependency and entity representations from auxiliary tasks into a more robust relation representation against the noisy labels. The experimental results demonstrate that our model improves the predictive performance substantially over single-task learning baselines.

Summary With the thriving of the social media (eg, Sina Microblog and Twitter), the public is keen on expressing their opinions or views on entities as celebrities and products. Emotion mining on social media can be applied in diverse areas, such as helping government or organizations understand people's attitudes so as to make right decisions for business services or political campaigns. One kind of the mainstream approaches for emotion mining are based on topic models, while most of these existing approaches aim at analyzing entire sentiments for the whole Microblog, rather than explicitly assigning the relevant sentiments to the specific entities. In addition, topic model is mainly based on the bag‐of‐words model but ignores the semantic relations of entity‐word in the modeling process, which brings low accuracy and poor interpretability to the sentiment analysis. To overcome the aforementioned difficulties, an Entity Sentiment Topic Model (ESTM) is proposed, which carries out entity‐dependent sentiment analysis. To improve the accuracy of sentiment analysis and enhance the interpretability of the results, ESTM is integrated with relations of entity‐word and a six‐dimensional emotion lexicon as weakly supervised information. Experiments have shown promising results on sentiment classification accuracy, interpretability, and quality of coherent topics for entities.

Summary Users of social media have different influences on the evolution of a Web event. Finding influential users could benefit such information services as recommendation and market analysis. However, most of the existing methods are only based on social networks of users or user behaviors while the role of the contents contributed by users in social media is ignored. In fact, a Web event evolves with both user behaviors and the contents. This paper proposes an approach to find influential users by extracting user behavior network and association network of words within the contents and then uses PageRank algorithm and HITS algorithm to calculate the influence of users on the integration of two networks. The proposed approach is effective on several real‐world datasets.

A public safety event is a danger and urgent event that need early detection, quick response, and accuracy recover. The efficient method for responding to a happening public safety event is to collect and describe the related data. Besides the surveillance cameras from the physical space, the social media data can also be used to collect and describe the related data of a public safety event. In this work, the proposed method focuses on the step for describing public safety events. Given a public safety event, videos from the surveillance cameras and social messages from social sensors are collected. The multi-modal information including texts, images, videos, and spatial-temporal data is mined to give a description precisely and concisely. First, the social sensors are associated to surveillance cameras by the spatial and temporal information. In the second stage, the social messages are associated to surveillance cameras by the semantic information. In the third stage, the social messages are associated to surveillance cameras by the visual feature. Besides the text, social sensors may upload images or videos. Finally, the multi-modal description step is given based on the three different associations. The experiments on the real data demonstrate the superiority of the proposed framework. Case studies on the real public safety event show the proposed model has good performance and high effectiveness.

Summary A corpus (eg, patents or news texts) is an important knowledge resource that contains various topics, such as specific technologies or social events. Topic detection models of corpus, eg, Latent Dirichlet Allocation and KeyGraph, provide an important basis for exploring the status quo and trends in science, technology, or social events. However, these models suffer from low retrieval performance as they only consider text own explicit semantics in a single‐domain corpus. In addition, many incremental models, such as online‐LDA, are based on time slices. In this paper, a new topic detection model is proposed to improve the topic detection performance of a single‐domain corpus, which is inspired by a human memory cognitive process (THC). First, to improve the accuracy, distributions over words and inter‐word relations across a corpus are utilized as background knowledge, which is a type of implicit semantics, and we can find a more semantic‐sensitive part of texts. Second, to realize online topic detection without time slices, we introduce a probability gain‐based dynamic probabilistic model to detect latent topics by learning a model based on the dynamic human memory cognitive process. These two steps constitute the framework of our model. The experimental results for four public datasets (Reuters‐R8, Reuters‐R52, WebKB, and Cade12) reveal that our model is approximately ten percent higher than other baselines (eg, KeyGraph and LDA) on the Adjusted Rand Index (ARI).

An emergency event is a sudden, urgent, usually unexpected incident or occurrence that requires an immediate reaction or assistance for emergency situations, which plays an increasingly important role in the global economy and in our daily lives. Recently, the web is becoming an important event information provider and repository due to its real-time, open, and dynamic features. In this paper, web resources based states detecting algorithm of an event is developed in order to let the people know of an emergency event clearly and help the social group or government process the emergency events effectively. The relationship between web and emergency events is first introduced, which is the foundation of using web resources to detect the state of emergency events imaged on the web. Second, five temporal features of emergency events are developed to provide the basis for state detection. Moreover, the outbreak power and the fluctuation power are presented to integrate the above temporal features for measuring the different states of an emergency event. Using these two powers, an automatic state detecting algorithm for emergency events is proposed. In addition, heuristic rules for detecting the states of emergency event on the web are discussed. Our evaluations using real-world data sets demonstrate the utility of the proposed algorithm, in terms of performance and effectiveness in the analysis of emergency events.

Sparse nonnegative matrix factorization (SNMF) aims to factorize a data matrix into two optimized nonnegative sparse factor matrices, which could benefit many tasks, such as document-word co-clustering. However, the traditional SNMF typically assumes the number of latent factors (i.e., dimensionality of the factor matrices) to be fixed. This assumption makes it inflexible in practice. In this paper, we propose a doubly sparse nonparametric NMF framework to mitigate this issue by using dependent Indian buffet processes (dIBP). We apply a correlation function for the generation of two stick weights associated with each column pair of factor matrices while still maintaining their respective marginal distribution specified by IBP. As a consequence, the generation of two factor matrices will be columnwise correlated. Under this framework, two classes of correlation function are proposed: 1) using bivariate Beta distribution and 2) using Copula function. Compared with the single IBP-based NMF, this paper jointly makes two factor matrices nonparametric and sparse, which could be applied to broader scenarios, such as co-clustering. This paper is seen to be much more flexible than Gaussian process-based and hierarchial Beta process-based dIBPs in terms of allowing the two corresponding binary matrix columns to have greater variations in their nonzero entries. Our experiments on synthetic data show the merits of this paper compared with the state-of-the-art models in respect of factorization efficiency, sparsity, and flexibility. Experiments on real-world data sets demonstrate the efficiency of this paper in document-word co-clustering tasks.

Summary How to generate summary with more novel and rich semantics is a challenging issue in the area of multi‐document automatic summary. In this paper, a core semantics extraction model (CSEM) is proposed to improve the novel and rich semantics of multi‐document summary. Firstly, for improving the rich semantics, semantic units, which are a group of association relations of keywords, are used to express texts' semantics. Secondly, for improving the novel semantics, an attenuation function is introduced to adjust the importance of semantic units according to the appearing times that semantic units in the candidate of summary sentences. Thirdly, in order to maximize the novel and rich semantics of summary, the generating process of summary is converted into the optimization process on how to find a set of sentences with a higher importance. Finally, CSEM extracts the least number of sentences to cover the most core semantics in corpus as summary. Experimental results on the benchmark DUC 2004 show that our model outperforms the state‐of‐art approaches (eg, OCCAMS_V, JS‐Gen‐2) under official metric. Especially, the recall of our model in ROUGE‐1 is 40.684%, which is better than other approaches (eg, OCCAMS_V 38.497% and JS‐Gen‐2 36.739%).

Association link network (ALN) can organize massive Web information to provide many intelligent services in our big data society. Effective semantic layered technologies not only can provide theoretical support for knowledge discovery in Web resources, but also can improve the searching efficiency of related information systems such as Web information system and industrial information system. How to realize the layer division of association semantic by the hierarchy analysis of ALN is an important research topic. To solve this problem, this paper proposes a hierarchy-cutting model of association semantic. First, experiments of four types of keywords with different linking roles are conducted to discover the possible distribution law. Experimental results show that these keywords with association role reveal previous power-law distribution. Then, based on the discovered power-law distribution, up-cutting and down-cutting points are presented to divide the association semantic into three layers. At the same time, theories of the hierarchy-cutting model are presented. Finally, examples of current core topic and permanent topics belonging to a domain are given. The experiments show that hierarchy-cutting points have high accuracy. The multilayer theory of association semantic can provide a theoretical support for knowledge recommendation with different particle sizes on ALNs.

Traditional relational topic models provide a successful way to discover the hidden topics from a document network. Many theoretical and practical tasks, such as dimensional reduction, document clustering, and link prediction, could benefit from this revealed knowledge. However, existing relational topic models are based on an assumption that the number of hidden topics is known a priori, which is impractical in many real-world applications. Therefore, in order to relax this assumption, we propose a nonparametric relational topic model using stochastic processes instead of fixed-dimensional probability distributions in this paper. Specifically, each document is assigned a Gamma process, which represents the topic interest of this document. Although this method provides an elegant solution, it brings additional challenges when mathematically modeling the inherent network structure of typical document network, i.e., two spatially closer documents tend to have more similar topics. Furthermore, we require that the topics are shared by all the documents. In order to resolve these challenges, we use a subsampling strategy to assign each document a different Gamma process from the global Gamma process, and the subsampling probabilities of documents are assigned with a Markov Random Field constraint that inherits the document network structure. Through the designed posterior inference algorithm, we can discover the hidden topics and its number simultaneously. Experimental results on both synthetic and real-world network datasets demonstrate the capabilities of learning the hidden topics and, more importantly, the number of topics.

With the advances of information communication technologies, it is critical to improve the efficiency and accuracy of modern data processing techniques. The past decade has witnessed the tremendous technical advances in sensor networks, Internet/Web of Things, cloud computing, mobile/embedded computing, spatial/temporal data processing, and big data, and these technologies have provided new opportunities and solutions to data processing techniques. Big data is an emerging paradigm applied to datasets whose size is beyond the ability of commonly used software tools to capture, manage, and process the data within a tolerable elapsed time. Such datasets are often from various sources (Variety) yet unstructured such as social media, sensors, scientific applications, surveillance, video and image archives, Internet texts and documents, Internet search indexing, medical records, business transactions, and web logs and are of large size (Volume) with fast data in/out (Velocity). More importantly, big data has to be of high value (Value) and establish trust in it for business decision-making (Veracity). Various technologies are being discussed to support the handling of big data such as massively parallel processing databases, scalable storage systems, cloud computing platforms, and MapReduce. Big data is more than simply a matter of size; it is an opportunity to find insights in new and emerging types of data and content, to make business more agile, and to answer questions that were previously considered beyond our reach. This special issue wants to demonstrate the emerging issues in the research of big data and approaches towards it. Original and research articles are solicited in all aspects including theoretical studies, practical applications, and experimental prototypes. The submitted manuscripts were reviewed by experts from both academia and industry. After two rounds of reviewing, the highest quality manuscripts were accepted for this special issue. Totally, we have received 20 manuscripts and 11 papers are accepted. Five papers are selected from SKG2015 conference with about 50% new content. This special issue will be published by Concurrency and Computation: Practice and Experience as special issues. In order to detect and describe the real-time urban emergency event, the knowledge base model is proposed by the paper by Z. Xu et al 1. The crowdsourcing-based knowledge base model is introduced, which uses the information from social media. X. Lin et al 2. introduced one such comprehensive schemes: it takes consideration of the DG's own characteristics and its ability to support the local loading, it adopts different protection strategy, and realizes the fault isolation and island division through the coordination between the protection and the automatic devices. J. Zheng 3 extracted the feature information and also mined the association rules to identify and mark the unknown protocol using the learning mechanism of machine. The unknown protocol in a specific environment was found and analyzed by marking the fingerprint information of protocol. M. Xie 4 studied meta-analytically integrated results from piles of network position researches in organizational innovation in 20 years. Considering the real social community network partition approach regardless of the directed and weighted characteristic, K. Gao 5 proposed a novel algorithm in pervasive sensing environment. Start with classifying and comparing current search engines, particularly from the perspective of search patterns which consist of index structure, user profiles, and interaction mechanism. X. Wei 6 then present a novel search pattern named ExNa by defining its model and basic operations in detail. Two attribute reduction methods based on minimum decision cost are proposed by Z. Bi 7 from the algebraic view and the information theory, respectively. Y. Wang 8 designs an optimization algorithm to solve the problem. Numerical experiments show that the optimization algorithm can solve the load imbalance of the METGRID to some extent, and the computation speed of the METGRID and REAL modules after the optimization on 64 cores is about 7.2 times faster than before. Y. Xue 9 proposed Event Space Model for event analysis by multi-viewer. In the model, each event is mapped into Network Public Opinion Data Space (OS) and Actual Behavior Data Space (BS). S. Zhang 10 proposed a model for estimating the out-degree of any one node in ALN from semantic feature view, which can greatly reduce the searching scope for the rapid positioning of Web resources stored in large-scale database. J. Zhang 11 presented how to mine patent text to get valuable information/knowledge from large-scale candidates obtained from these patents based on massive patent texts. The guest editors would like to thank Prof. Geoffrey C. Fox, who is the editor in chief of Concurrency and Computation: Practice and Experience. His help and trust are the most important thing for the success of this special issue. The guest editors would like to thank the reviewers for their high-quality reviews, which provided insightful and constructive feedback to the authors of the papers.

The evolution potential estimation of news events can support the decision making of both corporations and governments. For example, a corporation could manage its public relations crisis in a timely manner if a negative news event about this corporation is known with large evolution potential in advance. However, existing state-of-the-art methods are mainly based on time series historical data, which are not suitable for the news events with limited historical data and bursty properties. In this article, we propose a purely content-based method to estimate the evolution potential of the news events. The proposed method considers a news event at a given time point as a system composed of different keywords, and the uncertainty of this system is defined and measured as the Semantic Uncertainty of this news event. At the same time, an uncertainty space is constructed with two extreme states: the most uncertain state and the most certain state. We believe that the Semantic Uncertainty has correlation with the content evolution of the news events, so it can be used to estimate the evolution potential of the news events. In order to verify the proposed method, we present detailed experimental setups and results measuring the correlation of the Semantic Uncertainty with the Content Change of news events using collected news events data. The results show that the correlation does exist and is stronger than the correlation of value from the time-series-based method with the Content Change. Therefore, we can use the Semantic Uncertainty to estimate the evolution potential of news events.

Webpage recommendations for hot Web events can assist people to easily follow the evolution of these Web events. At the same time, there are different levels of semantic uncertainty underlying the amount of Webpages for a Web event, such as recapitulative information and detailed information. Apparently, the grasp of the semantic uncertainty of Web events could improve the satisfactoriness of Webpage recommendations. However, traditional hit-rate-based or clustering-based Webpage recommendation methods have overlooked these different levels of semantic uncertainty. In this paper, we propose a framework to identify the different underlying levels of semantic uncertainty in terms of Web events, and then utilize these for Webpage recommendations. Our idea is to consider a Web event as a system composed of different keywords, and the uncertainty of this keyword system is related to the uncertainty of the particular Web event. Based on keyword association linked network Web event representation and Shannon entropy, we identify the different levels of semantic uncertainty, and construct a semantic pyramid (SP) to express the uncertainty hierarchy of a Web event. Finally, an SP-based Webpage recommendation system is developed. Experiments show that the proposed algorithm can significantly capture the different levels of the semantic uncertainties of Web events and it can be applied to Webpage recommendations.

Incorporating the side information of text corpus, i.e., authors, time stamps, and emotional tags, into the traditional text mining models has gained significant interests in the area of information retrieval, statistical natural language processing, and machine learning. One branch of these works is the so-called Author Topic Model (ATM), which incorporates the authors's interests as side information into the classical topic model. However, the existing ATM needs to predefine the number of topics, which is difficult and inappropriate in many real-world settings. In this paper, we propose an Infinite Author Topic (IAT) model to resolve this issue. Instead of assigning a discrete probability on fixed number of topics, we use a stochastic process to determine the number of topics from the data itself. To be specific, we extend a gamma-negative binomial process to three levels in orderto capture the author-document-keyword hierarchical structure. Furthermore, each document is assigned a mixed gamma process that accounts for the multi-author's contribution towards this document. An efficient Gibbs sampling inference algorithm witheach conditional distribution being closed-form is developed for the IAT model. Experiments on several real-world datasets show the capabilities of our IAT model to learn the hidden topics, authors' interests on these topics and the number of topics simultaneously.

Summary Online hot event discovery has become a flourishing frontier where online document streams are monitored to discover newly occurring events or assigned to previously detected events. However, hot events have the nature to evolve, and their inherent topically related words are also likely to evolve. It makes event discovery a challenging task for traditional‐mining approaches. Combining word association and semantic community, Association Link Network (ALN) organizes the loosely distributed associated resources. This paper presents an ALN‐based novel online hot event discovery approach. Technically, this approach is enacted around three stages. In the first stage, we extract significant features to represent the content of each document from the online document stream. During the second stage, we classify the online document stream into topically related detected events considering event evolution in the form of ALN. At the third stage, we create an ALN‐based event detection algorithm, which is used to timely discover newly occurring hot events. The online datasets used in our empirical studies are acquired from Baidu News, which spans a range of 1315 hot events and 236,300 documents. Experimental results demonstrate the hot events discovery ability with respect to high accuracy, good scalability, and short runtime. Copyright © 2014 John Wiley & Sons, Ltd.

Graph mining has been a popular research area because of its numerous application scenarios. Many unstructured and structured data can be represented as graphs, such as, documents, chemical molecular structures, and images. However, an issue in relation to current research on graphs is that they cannot adequately discover the topics hidden in graph-structured data which can be beneficial for both the unsupervised learning and supervised learning of the graphs. Although topic models have proved to be very successful in discovering latent topics, the standard topic models cannot be directly applied to graph-structured data due to the "bag-of-word" assumption. In this paper, an innovative graph topic model (GTM) is proposed to address this issue, which uses Bernoulli distributions to model the edges between nodes in a graph. It can, therefore, make the edges in a graph contribute to latent topic discovery and further improve the accuracy of the supervised and unsupervised learning of graphs. The experimental results on two different types of graph datasets show that the proposed GTM outperforms the latent Dirichlet allocation on classification by using the unveiled topics of these two models to represent graphs.

Online comment has become a popular and efficient way for sellers to acquire feedback from customers and improve their service quality. However, some key issues need to be solved about evaluating and improving the hotel service quality based on online comments automatically, such as how to use the less trustworthy online comments, how to discover the quality defects from online comments, and how to recommend more feasible or economical evaluation indexes to improve the service quality based on online comments. To solve the above problems, this paper first improves fuzzy comprehensive evaluation (FCE) by importing trustworthy degree to it and proposes an automatic hotel service quality assessment method using the improved FCE, which can automatically get more trustworthy evaluation from a large amount of less trustworthy online comments. Then, the causal relations among evaluation indexes are mined from online comments to build the fuzzy cognitive map for the hotel service quality, which is useful to unfold the problematic areas of hotel service quality, and recommend more economical solutions to improving the service quality. Finally, both case studies and experiments are conducted to demonstrate that the proposed methods are effective in evaluating and improving the hotel service quality using online comments.

Similarity measures are the foundations of many research areas, e.g. information retrieval, recommender system and machine learning algorithms. Promoted by these application scenarios, a number of similarity measures have been proposed and proposing. In these state-of-the-art measures, vector-based representation is widely accepted based on Vector Space Model (VSM) in which an object is represented as a vector composed of its features. Then, the similarity between two objects is evaluated by the operations on two corresponding vectors, like cosine, extended jaccard, extended dice and so on. However, there is an assumption that the features are independent of each others. This assumption is apparently unrealistic, and normally, there are relations between features, i.e. the co-occurrence relations between keywords in text mining area. In this paper, a space geometry-based method is proposed to extend the VSM from the orthogonal coordinate system (OVSM) to affine coordinate system (AVSM) and OVSM is proved to be a special case of AVSM. Unit coordinate vectors of AVSM are inferred by the relations between features which are considered as angles between these unit coordinate vectors. At last, five different similarity measures are extended from OVSM to AVSM using unit coordinate vectors of AVSM. Within the numerous application fields of similarity measures, the task of text clustering is selected to be the evaluation criterion. Documents are represented as vectors in OVSM and AVSM, respectively. The clustering results show that AVSM outweighs the OVSM.

SUMMARY Association relations between concepts are a class of simple but powerful regularities in binary data, which play important roles in enterprises and organizations with huge amounts of data. However, although there can be easily large number of association relation mined from databases, since existing objective and subjective methods scarcely take semantics into consideration, it has been recognized early in the knowledge discovery literature that most of them are of no interest to the user. In this paper, the semantic discrimination capability (SDC) of association relation is measured based on discrimination value model first. The formula of SDC integrating both statistical and graph features is proposed from five different strategies. The high correlation coefficient of the proposed method against discrimination value shows that the proposed SDC measure is accuracy. Moreover, an application using SDC on document clustering is carried out, which shows that SDC has broad prospects on data‐related task such as document clustering. Copyright 2013 John Wiley © Sons, Ltd.

This paper advocates for a novel approach to recommend texts at various levels of difficulties based on a proposed method, the algebraic complexity of texts (ACT). Different from traditional complexity measures that mainly focus on surface features like the numbers of syllables per word, characters per word, or words per sentence, ACT draws from the perspective of human concept learning, which can reflect the complex semantic relations inside texts. To cope with the high cost of measuring ACT, the Degree-2 Hypothesis of ACT is proposed to reduce the measurement from unrestricted dimensions to three dimensions. Based on the principle of “mental anchor,” an extension of ACT and its general edition [denoted as extension of text algebraic complexity (EACT) and general extension of text algebraic complexity (GEACT)] are developed, which take keywords' and association rules' complexities into account. Finally, using the scores given by humans as a benchmark, we compare our proposed methods with linguistic models. The experimental results show the order GEACT>EACT>ACT> Linguistic models, which means GEACT performs the best, while linguistic models perform the worst. Additionally, GEACT with lower convex functions has the best ability in measuring the algebraic complexities of text understanding. It may also indicate that the human complexity curve tends to be a curve like lower convex function rather than linear functions.

How to build a text knowledge representation model, which carries rich knowledge and has a flexible reasoning ability as well as can be automatically constructed with a low computational complexity, is a fundamental challenge for reasoning-based knowledge services, especially with the rapid growth of web resources. However, current text knowledge representation models either lose much knowledge [e.g., vector space model (VSM)] or have a high complex computation [e.g., latent Dirichlet allocation (LDA)]; even some of them cannot be constructed automatically [e.g., web ontology language, (OWL)]. In this paper, a novel text knowledge representation model, power series representation (PSR) model, which has a low complex computation in text knowledge constructing process, is proposed to leverage the contradiction between carrying rich knowledge and automatic construction. First, concept algebra of human concept learning is developed to represent text knowledge as the form of power series. Then, degree-2 power series hypothesis is introduced to simplify the proposed PSR model, which can be automatically constructed with a lower complex computation and has more knowledge than the VSM and LDA. After that, degree-2 power series hypothesis-based reasoning operations are developed, which provide a more flexible reasoning ability than OWL and LDA. Furthermore, experiments and comparisons with current knowledge representation models show that our model has better characteristics than others when representing text knowledge. Finally, a demo is given to indicate that PSR model has a good prospect over the area of web semantic search.

As a free online encyclopedia with a large-scale of knowledge coverage, rich semantic information and quick update speed, Wikipedia brings new ideas to measure semantic correlation. In this paper, we present a new method for measuring the semantic correlation between words by mining rich semantic information that exists in Wikipedia. Unlike the previous methods that calculate semantic relatedness merely based on the page network or the category network, our method not only takes into account the semantic information of the page network, also combines the semantic information of the category network, and it improve the accuracy of the results. Besides, we analyze and evaluate the algorithm by comparing the calculation results with famous knowledge base (e.g., Hownet) and traditional methods based on Wikipedia on the same test set, and prove its superiority.

No abstract available.

With the rapid development of the Web, information on the Web, especially for textual information, has come to an extremely large amount and increases very fast every day. It has become much more difficult for people to find their interested or demanded information from the Web. Thus how to automatically and intelligently acquire the textual semantics from the Web has become an increasingly important issue today. In this paper, based on human memory and reading process, we propose a machine-oriented text understanding framework, which can guide machines to read texts with human ways so as to effectively acquire the semantics in texts and then further help people to find out the proper information they want. The proposed framework is on the basis of different kinds of memory systems in cognitive psychology and accordingly consists of four main parts, including shallow semantics acquiring model (SSAM), shallow-deep semantics processing model (SDSPM), episodic semantics activating model (ESAM) and background knowledge activating model (BKAM). A case study has been carried out to further specify the proposed framework.

With the increase of information scale of web events on the time, it is extremely difficult and challenging to grasp the semantics of web events artificially, because of the limitation of the time and energy of human beings. Herein, we propose a method to map the web event to keyword level association link network (KALN) for deep analysis of the semantics of web events, such as the evolution semantics of web events. Firstly, the original KALN is constructed at a given time by traditional data mining technologies. Then, the hierarchical KALN, consisted of Theme Layer Network, Backbone Layer Network and Tidbit Layer Network, is built based on the original KALN by information entropy to identify the different semantic levels of the web event, including stable semantics, sub-stable semantics and unstable semantics. With the semantic analysis of hierarchical KALN, human could easily gain a thorough understanding of the web event. Finally, experiments show that our method can effectively capture the different level semantics of web events.

Web search is a key information retrieval method for human beings in current society, both in the fields of academic and commercial activities. Due to "one-size-fits-all" approach limit to search results obtainment, there are still challenges to give personalized web service with high precision in traditional web search process. Herein, a new framework is proposed to advance traditional search process into a new paradigm, i.e. interactive search. First, a cognitive model of interaction process about one search step is developed by the imitation of human search behaviors. Then based on the theory of interactive computing, an interactive search model is introduced to formalize successive search sessions which include several search steps. Third, based on human's cognitive mechanism such as the spreading activation model and user memory theory, a user model is designed to capture user's search activities which can aid the interactive search process effectively. Last, by the help of Associated Link Network, an information gradient based rank algorithm is proposed aiming at maximizing the quantity of web information supply. The efficiency of our proposed interactive search service is verified by the experimental results.

The performance of a service's QoS may evolve relatively frequently with its internal changes or the changes of dynamic Internet environment, especially when some"intentional" deceptions are taken into consideration. Therefore, the service providers could not always deliver their services according to their "promised" quality. In view of this challenge, a history record-based service optimization method, named Hire Some, is investigated in this paper. This method aims at enhancing the credibility of service composition plan, taking advantage of a web service's QoShistory records, rather than using the tentative QoS values advertised by the service provider. At last, a case study and an experiment is presented for validating the method.

Abstract Queries to Web search engines are usually short and ambiguous, which provides insufficient information needs of users for effectively retrieving relevant Web pages. To address this problem, query suggestion is implemented by most search engines. However, existing methods do not leverage the contradiction between accuracy and computation complexity appropriately (e.g. Google's ‘Search related to’ and Yahoo's ‘Also Try’). In this paper, the recommended words are extracted from the search results of the query, which guarantees the real time of query suggestion properly. A scheme for ranking words based on semantic similarity presents a list of words as the query suggestion results, which ensures the accuracy of query suggestion. Moreover, the experimental results show that the proposed method significantly improves the quality of query suggestion over some popular Web search engines (e.g. Google and Yahoo). Finally, an offline experiment that compares the accuracy of snippets in capturing the number of words in a document is performed, which increases the confidence of the method proposed by the paper. Copyright © 2010 John Wiley & Sons, Ltd.

SUMMARY Semantic similarity measures play important roles in many Web‐related tasks such as Web browsing and query suggestion. Because taxonomy‐based methods can not deal with continually emerging words, recently Web‐based methods have been proposed to solve this problem. Because of the noise and redundancy hidden in the Web data, robustness and accuracy are still challenges. In this paper, we propose a method integrating page counts and snippets returned by Web search engines. Then, the semantic snippets and the number of search results are used to remove noise and redundancy in the Web snippets (‘Web‐snippet’ includes the title, summary, and URL of a Web page returned by a search engine). After that, a method integrating page counts, semantics snippets, and the number of already displayed search results are proposed. The proposed method does not need any human annotated knowledge (e.g., ontologies), and can be applied Web‐related tasks (e.g., query suggestion) easily. A correlation coefficient of 0.851 against Rubenstein–Goodenough benchmark dataset shows that the proposed method outperforms the existing Web‐based methods by a wide margin. Moreover, the proposed semantic similarity measure significantly improves the quality of query suggestion against some page counts based methods. Copyright © 2011 John Wiley & Sons, Ltd.

Association Link Network (ALN) aims to establish associated relations among various resources. By extending the hyperlink network World Wide Web to an association-rich network, ALN is able to effectively support Web intelligence activities such as Web browsing, Web knowledge discovery, and publishing, etc. Since existing methods for building semantic link on Web resources cannot effectively and automatically organize loose Web resources, effective Web intelligence activities are still challenging. In this paper, a discovery algorithm of associated resources is first proposed to build original ALN for organizing loose Web resources. Second, three schemas for constructing kernel ALN and connection-rich ALN (C-ALN) are developed gradually to optimize the organizing of Web resources. After that, properties of different types of ALN are discussed, which show that C-ALN has good performances to support Web intelligence activities. Moreover, an evaluation method is presented to verify the correctness of C-ALN for semantic link on documents. Finally, an application using C-ALN to organize Web services is presented, which shows that C-ALN is an effective and efficient tool for building semantic link on the resources of Web services.

Similar Link Network (SiLN) is a semantic over layer on Web resources with similar relations instead of hyperlinks, which aims at providing plentiful semantics for intelligent Web activities. However, SiLN is difficult to be built based on cosine computation in large-scale Web resources due to its high building time complexity and weak connectivity. Herein, three strategies are proposed to address those issues. First, dividing and conquering strategy is applied to divide the large-scale Web resources into amounts of rough similar communities, which reduces SiLN's building time complexity significantly. After that, a multi-level structure network is designed to effectively manage the large-scale Web resources to guarantee SiLN's connectivity. Finally, two-level feedback with isolated resources strategy is developed to improve the accuracy of the building of SiLN. Experimental results have proved that our proposed method of building SiLN is feasible and efficient, with the merits of low complexity, good connectivity and high precision.

Abstract The semantically associated network on the Web is a Semantic Link Network built by mining the associated relation between Web pages. The associated link from page A to page B indicates that users who have browsed page A is likely to also browse page B. This paper explores the statistical properties of the associated network on the Web. Web pages of a specific domain are automatically downloaded by a Web crawler to build an associated network. We analyze the associated network at different domain thresholds and classify the topology into three states, that is, the original state, the kernel state and the final state. A mathematical model is built to study the in‐degree distribution, the out‐degree distribution and the total‐degree distribution for both the kernel state and the final state. By tuning the model parameters to reasonable values, we obtain the distinct power‐law forms for the three degree distributions with exponents that agree well with the statistical data. The proposed model can not only describe the evolving processes of the associated network on the Web, but also provides theory basis for complex applications such as semantic community discovery, intelligent browsing and recommendation. Copyright © 2009 John Wiley & Sons, Ltd.

Malignant tumor or cancer is one of serious diseases threatening human health. It is a challenging issue on how to diagnose the malignant tumor earlier and more accurately. A feasible way on cancer diagnosis is efficiently to share and analyze tumor related data. We work on a cyberinfrastructure which presents data sharing and data analysis solutions to reveal the relationship between tumor and gene.

Association link network (ALN) is used to establish associated relations among various resources, aiming at extending the hyperlink network World Wide Web to an association-rich network, for effectively supporting Web intelligence activities. Unfortunately, with the increase number of Web resources, the challenge of incremental building of ALN is on how to perform the association weight of the new coming Web resources efficiently and exactly. A naive way is to compare every pair of data in the existing ALN, thus bearing a O(n2) time complexity. Given the scale of the Web, it is unrealistic to compute the association weight between the new coming Web data and each data in the existing ALN, respectively. In this paper, a new method based on All-Pairs algorithm for incremental building of ALN is proposed. The experiments and evaluations show that our incremental building method performs a high accuracy. Moreover, the scale-independent property of our method make it more appropriate to be used on the Web.

Key-word based researches of service discovery focus on direct match of userpsilas requirements and often neglect relations between services. While techniques based on conventional semantic offer many kinds of relations, considerable time is spent on reasoning. In this paper, we utilizes E-FCM (Element Fuzzy Cognitive Map) to describe services for the reason that E-FCM can keep the semantic information as much as possible and E-FCMs can be automatically created for web services. Furthermore, instead of reasoning, the semantic relations among E-FCM are built based on computation, therefore semantic relations among services can be found out quickly. We focus on the associated semantic relations among services because complex applications always comprise of services with associated functions. The associated link network (ALN) is constructed upon associated relations to generate associated web service flows, which can be used to create complex applications, thus to facilitate discovery efficiency and improve utilization of services.

Personalized services provide specific services to satisfy userpsilas real-time requirements. How to effectively discover and organize proper Web resources is a key issue. Based on Web knowledge flow which is used to semantically organize and represent Web resources that are recommended to user, this paper presents a 3-way strategy of finding proper resources for a user in multi-user environment. With this strategy, generation of Web Knowledge flow can be accomplished based on collaborative user and semantics link network. First, according to the similarity degree between active user and collaborative user, collaborative user is refined to strict kind and loose kind. Secondly, userpsilas browsing sequence of Web resources is presented to find collaborative users and implemented in a 3-D coordinate space. Thirdly, average information entropy of semantic relationship between Web resources is presented to evaluate the similarity of userspsila browsing feature. The experimental results demonstrate the validity of the method. It can be seen that the proposed method has a brilliant perspective in the applications of Web personalized services.

Abstract Similarity Knowledge Flow (SKF) is a kind of scientific workflow, providing an effective technique and theoretical support for intelligent browsing in the Web and e‐Science environment. In this paper, a Semantic Link Networks (SLN) based SKF generation method is proposed. First, the topics are represented by Element Fuzzy Cognitive Maps then the semantic values of concepts/keywords and relations are calculated. Third, semantic similarity degrees between topics are calculated to build SLN‐based semantic values of concepts and their relations in Element Fuzzy Cognitive Maps. In this way, similar relations at the keyword level are extended to the topic level. With the help of SLN and based on user's demand, SKF is generated as the browsing path of topics to guide user browsing behaviors. Finally, the semantic value of SKF is defined as a criterion to evaluate the browsing path of topics. Experimental results show that the browsing path of topics is easy to be activated by SKF which is generated by SLN. The proposed method has been proved to have a very good prospect in the fields of Web services and e‐Science applications. Copyright © 2009 John Wiley & Sons, Ltd.

Traditional text mining techniques have weak ability to provide associated relations with rich semantics that is a foundation of the intelligent browsing of topics, discovery of semantic community and precise personalized recommendation in current Web and knowledge Grid, etc. In this paper we propose an algorithm to generate and calculate the associated relations and their strengths between documents within a domain. Each document is represented by a bag of words and their weights. We first build domain knowledge background based on the association rules at keyword level, and then we apply those association rules to generate and calculate the documents' semantic relations and their strengths at document level, which effectively shorten the semantic gap from keyword semantics to document semantics. Experimental results show that our proposed method is feasible and able to discover interesting facts within a domain.

An intelligent model of relative words that concerns semantic similarity is discussed in this paper. Based on this model, a knowledge system capable of semantically calculating and associating has been developed, which enables adapting and/or extending keywords in search process. This paper aims to make information retrieval with semantic ability; as a result, the precision of information retrieval could be improved to some degree. In final, the knowledge system has been applied to the Digit Ocean Share Platform as an application presentation.

Communications among Web services are asynchronous. Asynchronous models of service compositions face the problem that the performance of verification is bring down with states explosion. We describe an approach, which utilize the interaction-independence of component services in the composition and forms all component services into different groups. Preliminary experiment results show that the verification of service compositions upon these groups can decrease the size of exploration states and hence improve the performance.

Abstract Semantic representation of a topic is the basic element representation of a scientific document, which can be reflected by keywords' relations and their weights, as well as the state values of co‐occurring keywords in a documental fragment in which a topic is discussed. Concept lattice and probability statistical methods are proposed to generate fuzzy cognitive maps (FCMs) as the original semantic representation of documental fragments (SRDFs) for the e‐science Knowledge Grid. The additive capability of FCMs is used to merge SRDFs for reducing the noise and redundant information hidden in the original SRDFs. The loss of documental information in the merging process of SRDFs is measured by the proposed semantic merging entropy. Based on the distributional property of keywords on SRDFs, semantic distribution entropy is introduced to evaluate the change of keywords' distribution on SRDFs with the merging process. The semantic representation of a scientific document is automatically generated by the interconnection of the merged SRDFs belonging to one domain. The generating, merging and evaluating methods of the semantic representation of a scientific document are validated by experiments. The proposed semantic representation highlights the relations among keywords and topics, which reduces the complexity of document analysis in an e‐science Knowledge Grid effectively. Copyright © 2007 John Wiley & Sons, Ltd.

Abstract Domain keywords of text play a primary role in text classifying, clustering and personalized services. This paper proposes a term frequency inverse document frequency (TFIDF) based method called TDDF (TFIDF direct document frequency of domain) to extract domain keywords from multi‐texts. First, we discuss the optimal parameters of TFIDF, which are used to extract textual keywords and domain keywords. Second, TDDF is proposed to extract domain keywords from multi‐texts, which takes document frequency of domain into account. Finally, the distribution of domain keywords on scientific texts is studied. Experiments and applications show that TDDF is more effective than the optimal TFIDF in the extraction of domain keywords. Domain keywords accord with normal distribution on a single text after deleting the ubiquitous domain keywords. Copyright © 2008 John Wiley & Sons, Ltd.

Abstract Textual knowledge flow (TKF) provides an effective technique and theoretical support for the discovery and cooperation of knowledge innovation, intelligent browsing, and personalized recommendation in Web services and e‐Science Knowledge Grid. For the discovery of TKF, firstly knowledge map (KM) is proposed to represent the textual knowledge; then a hash algorithm is used to code KMs in order to form an Island which contains enormous KMs belonging to a domain. Under the control of the Island, C‐Location and R‐Location are introduced to manage those KMs belonging to an Island. KM‐Chord is proposed to manage the number of Islands, C‐Locations and R‐Locations in Web or a library. With the help of the management of KMs, similar relation and associated relation between KMs are found to build the semantic link network (SLN) between KMs. Based on the SLN and users' profile and input, similar or associated TKF with the user's different demands is activated. Experiments show that the proposed method can effectively discover TKF for Web services and e‐Science Knowledge Grid. Copyright © 2008 John Wiley & Sons, Ltd.