Structuring Chinese recipes is inherently challenging due to their complex procedural logic and the semantic density of cooking instructions. This paper introduces the Chinese Recipe Flow Graph (CRFG), a novel dataset designed for fine-grained recipe modeling. Based on rigorous annotation protocols, we performed multi-round manual labeling of entities and relationxds within cooking texts. To bridge the gap between culinary actions and health metrics, Large Language Models (LLMs) were employed to map ingredients to the China Food Composition Table. Furthermore, we developed LogicFlow-Net, which utilizes Inter-sentence Position enhancement to capture long-range dependencies in complex procedural text. The resulting dataset comprises 304 recipes, 25,025 entities, 19,369 relation instances, and 727 ingredients with detailed nutritional data. Experimental results demonstrate that the proposed model achieves F1 scores of 82% and 91% for Named Entity Recognition (NER) and Relation Extraction (RE) tasks, respectively. Although the scale and diversity of the dataset require further expansion, this study achieves fine-grained modeling of recipe logic and nutritional data. Consequently, it provides high-quality data and methodological support for downstream applications such as intelligent cooking, nutrition management, and personalized recipe recommendation.

Few-shot molecular property prediction is a persisting challenge in many biology-related tasks, because the same molecule may exhibit different properties (e.g., active or inactive) in different tasks. Existing methods view all atoms as equally important and attend to predict the properties by averaging the features of similar molecules, which ignores key substructures within molecules and leads to poor prediction performance. Since a molecule implicitly includes key substructures, which determines the properties of the molecular, and the atom combination forms key substructures, we focus on the atom combination in this paper. With this, we propose the Hierarchical Molecular Attention Network (HMAN) to predict the molecular properties through combining atoms. First, we utilize the average pooling to extract both the prototype and the molecular features, and use Graph Neural Networks (GNN) to extract the atomic features, then concatenates the prototype and molecular features with all atomic features as input to the self-attention mechanism to calculate the different weights for atoms. Here, we select the top $B$ atoms with the highest attention scores to form the key substructures. Second, we view the formed key substructures as the query vectors and regard the molecular features as the key-value pairs, and then feed them into another self-attention to obtain the scores of key substructures. We still select $k$ key substructures with the highest scores, and weight the sum of a molecular feature and $k$ key substructures to predict the molecular properties. To train the HMAN, we design a new loss function, which includes Binary Cross-Entropy and a weighted negative log-likelihood. The former is to predict the molecular properties, and the last one is to optimize the weight distribution, rendering that the weight distribution matches the predicted probability distribution, with back-propagation. Theoretical analysis proves the convergence of the new loss function, and extensive experimental results demonstrate that HMAN significantly outperforms SOTA baseline models in molecular property prediction tasks.

Deep Learning (DL) has significantly advanced artificial intelligence (AI) across numerous applications. However, its foundational reliance on centralized data collection introduces critical limitations concerning privacy, security, and scalability. With the continued proliferation of the Internet of Things (IoT), massive volumes of sensitive data are generated at the network edge, necessitating collaborative learning systems capable of securely sharing information without compromising confidentiality. Federated Learning (FL) partially addresses these challenges by enabling on-device model training, but its dependence on a central coordinator remains a core vulnerability, creating communication bottlenecks, fairness constraints, and susceptibility to single-point-of-failure (SPOF) risks. Swarm Learning (SL), developed in collaboration with Hewlett Packard Enterprise (HPE), represents a decentralized advancement that mitigates these limitations by eliminating central orchestration. Through the integration of blockchain technology with distributed machine learning (ML), SL establishes a secure, transparent, and fault-tolerant paradigm for peer-to-peer model exchange and aggregation. This survey provides a comprehensive overview of the architectural foundations, enabling technologies, and applications of SL in sensitive domains, including healthcare, the Internet of Vehicles (IoV), and Industrial IoT. In addition, it introduces a comparative taxonomy that systematically categorizes existing research by scope, methodological approach, and evaluation maturity. The review concludes by identifying key research directions—such as lightweight consensus protocols, energy-efficient optimization, and cross-chain interoperability—to advance the development of practical, secure, and privacy-preserving SL systems.

Emergency events like traffic accidents and natural disasters frequently cause severe disruptions to urban traffic patterns, posing challenges to conventional forecasting methods based on historical data. Recent progress has explored integrating auxiliary textual information from social media, news reports, and incident details to enhance forecasting models. However, these approaches often fail to effectively align semantic context from textual data with the spatio-temporal patterns in urban traffic flow. To bridge this disparity, we propose a novel framework called TPGM-LLM, which leverages text-prompt-guided multimodal large language models to dynamically assimilate real-time data, including incident reports, traffic sensors, and weather conditions. Specifically, the proposed framework integrates a pre-trained LLM-based text encoder to interpret descriptions of emergency events as semantic prompts. Furthermore, it incorporates a dynamic spatio-temporal hypergraph module that employs FastDTW to capture non-local dependencies among different road segments. Additionally, a multimodal feature extraction LLM is utilized to merge textual guidance with traffic dynamics in a hierarchical manner, enhancing the accuracy of long-term traffic forecasting. Experimental results on the Beijing Text-Traffic dataset (BjTT) demonstrate that the proposed model outperforms existing methods, highlighting its effectiveness in complex and dynamic traffic situations. Our code is available athttps://github.com/luyaxuan/TPGM-LLM

Graph neural network (GNN) is a powerful approach for graph-based recommender systems. Recently, GNNs integrated with contrastive learning have achieved superior performance in recommendations due to their data augmentation techniques for sparse data. These methods are susceptible to noisy perturbations and are affected by data bias stemming from missing-not-at-random (MNAR) issues, which are common in implicit feedback scenarios. In this paper, we propose a Counterfactual Intervention Enhancing Graph Contrastive Learning (CIEGCL) paradigm. Our model applies counterfactual intervention in the positive sampling process for contrastive augmentation and generates enhanced facts (e.g., random perturbations) for contrast penalty, improving model generalization. Additionally, our dual contrastive channels separately handle click prediction and recommendation tasks to reduce the impact of data bias on the final recommendations. Empirical studies on five real-world datasets demonstrate the effectiveness of our model over the state-of-the-arts. Further theoretical and experimental analysis confirms the rationale behind CIEGCL. We release our code at https://github.com/huang-zi-jian/CIEGCL.

Tabular data is one of the most common data formats, and recent advancements in deep learning have driven significant progress in tabular data synthesis. However, the complexity of mixed-type distributions in tabular data forces existing synthesis methods to rely on lossy preprocessing, leading to significant logical inconsistencies in synthetic data. Moreover, most approaches lack flexible conditional control over the synthesis process. In this paper, we introduce a novel two-stage tabular data synthesis method LogiCoTab, which fundamentally mitigates logical inconsistencies and enhances controllability in synthetic data. In the logical awareness stage, we design a dual-module architecture to extract semantic features and uncover intricate inter-column logical relationships. In the controlled synthesis stage, we develop a diffusion-based generative model enabling highly flexible and precise conditional data generation. Extensive experiments show that LogiCoTab outperforms state-of-the-art methods on multiple datasets, significantly enhancing tabular data synthesis quality.

Data imbalance is one of the main reasons for the degradation of machine learning model performance. In supervised learning, class imbalance often leads to a shift in the decision boundary, causing minority class samples containing critical information to be misclassified. To address this issue, this study proposes an oversampling method based on Generative Adversarial Networks (GANs) to improve the classification performance on imbalanced data. We design a novel training loss function for generator to better capture the distributional characteristics of input samples, thereby generating higher-quality samples to construct a balanced training dataset. Subsequently, we train and test the model on multiple datasets. Experimental results show that, compared to existing oversampling methods, the proposed approach significantly improves the performance of classification algorithms on 20 real-world imbalanced datasets.

The cybersecurity threats targeting industrial control systems (ICS) are evolving with increasing sophistication. Addressing the detection blind spots in existing source code analysis techniques, this study reveals a dual security paradox arising from code sensitivity: privacy leakage risks caused by decompilation techniques and integrity verification deficiencies in reverse engineering. This paper investigates three critical challenges: (1) What are the component flow process and detection elements of ICS component source code? (2) How can high-performance and reliable tracing and traceability be provided for ICS component source code exceptions and routine detection? (3) How can privacy enhancement and trusted detection of ICS component source code with high sensitivity be achieved? This paper proposes a blockchain-integrated trusted detection framework for ICS (BCTD-ICS), delivering groundbreaking solutions: (1) Establishing a lifecycle circulation model that systematically maps component types, stakeholders, and detection parameters; (2) Developing a tripartite collaborative architecture (Blockchain-Identification Resolution Zero-knowledge proofs (ZKPs)), featuring a traceability mechanism with trusted identification codes (resolution efficiency: 40ms/105 queries) to eliminate decompilation-induced privacy risks; (3) Creating an industrial-oriented privacy enhancement system utilizing DBSCAN clustering for intelligent sampling (26% compression rate on BCN3D Moveo) and optimizing ZK-SNARK protocols through Shamir’s Secret Sharing, establishing a backdoor-resistant distributed parameter generation system (time delay increment < 100ms). Experimentally verified, our solution enables ICS component code detection supply-chain-wise without sensitive data leakage in real-world industries. This work establishes a novel trusted detection paradigm for ICS, advancing detection efficiency and credibility under strict privacy preservation requirements, meeting Industry 4.0 security demands.

For soft sensor modeling of multimode chemical processes, a common method is to build an individual model corresponding to each mode. However, certain individual mode models may not be reliable due to the imbalanced data across different modes. The soft sensor built for one mode with sufficient data exhibits suboptimal performance for other modes with insufficient data. To address this issue, a deep transfer learning method is introduced for soft sensor and a deep transfer spatial–temporal slow feature regression framework (STSFE) is proposed. In the framework, a Siamese network is employed for slow feature extraction. In addition, the encoder–decoder structure is embedded for input reconstruction verifying the effectiveness of slow features. However, the Siamese network fails to consider correlation of variables in quality prediction. To address this limitation, the Siamese network is designed with an embedded spatial–temporal attention mechanism to construct the STSFE model, and the spatial–temporal slow features are augmented with the historical quality variable for current quality prediction. The STSFE model is initially trained for the mode with sufficient data (source domain), and then transfer learning is employed to facilitate knowledge transfer from the source domain to the target domain (the mode with insufficient data) by reusing the lower level extraction part of STSFE. The effectiveness of the proposed method is validated through a benchmark sewage treatment case and a real chemical process.

Addressing the persistent challenge of learning from imbalanced datasets is crucial in advancing machine learning applications. Standard machine learning algorithms typically assume that the input data is balanced, and they often struggle to effectively learn the distribution of minority class data when dealing with imbalanced data. To address this, our study designed an improved Generative Adversarial Networks (GANs) model, named MDGAN, for tabular sample synthesis to augment samples and balance the data distribution. MDGAN employs a multi-generator and multi-discriminator structure to capture non-connected subspace manifolds, thereby better fitting the complete data distribution. To enhance the diversity among the multiple generators, an exclusive loss among generators was designed, ensuring that each generator produces data of different modalities. Additionally, a contrastive loss was introduced to ensure that the generated samples better fit the minority class distribution and are separated from the majority class distribution, preventing blurred classification boundaries. Qualitative and quantitative tests were conducted on 25 real datasets, and the experimental results indicate that MDGAN outperforms traditional classical models and current advanced oversampling models.

Infrared small target detection (IRSTD) based on deep learning has received extensive research and application. However, deep learning models require a large amount of data to perform well, and the collection and standardization of infrared small target data is challenging, limiting the applicability of such models. To address this issue, this study proposes a data augmentation scheme for infrared small targets based on Generative Adversarial Networks (GANs). The proposed method is a two-step approach: the first step is the generation of clean backgrounds, and the second is the adaptive fusion of targets and backgrounds. In the background generation stage, we first use the Fast Marching Method (FMM) to fill background targets and obtain clean backgrounds. Then, we design a multi-generator and multi-discriminator GAN model (MGD-GAN) to generate high-quality and diverse background images. In the adaptive target-background fusion stage, we propose a dual-discriminator GAN network (FusionGAN), which allows the target mask to be adaptively fused with the background pixels. By combining real targets with generated backgrounds, new infrared small target images are generated, achieving the goal of data augmentation. Experiments conducted across three different scenarios demonstrate that the proposed data augmentation scheme effectively enhances the performance of both traditional and advanced detection models.

Alignment between the food images and the corresponding recipes is an emerging cross-modal representation learning task. In this task, the recipes are composed of three components, i.e., food title, ingredient lists, and cooking instructions, which require a fine-grained alignment between the features of the two modalities. Existing methods usually aggregate the recipes into global embeddings and then align them with the global image embeddings. Meanwhile, semantic classification is frequently used in these methods to regularize the embeddings of the two modalities. While these methods are efficient, there remain two problems: (1) Forcing the alignment between the global images and recipes embeddings may result in losing the component-specific information. (2) The high diversity of food appearance leads to high uncertainty in the semantic classification of food images and recipes. To solve these problems, we propose a Fine-grained Prompting and Alignment (FPA) model to enhance the feature extraction and bring more component-specific information for fine-grained alignment. Furthermore, to regularize the semantic information contained in the cross-modal features, we design an Evidential Semantic Consistency (ESC) loss to keep the cross-modal semantic consistency. We have conducted comprehensive experiments on the benchmark dataset Recipe1M and the state-of-the-art results on the cross-modal recipe retrieval task demonstrate the effectiveness of our method.

Zero-shot learning (ZSL) aims to predict classes that have never appeared in the training stage. The key to addressing those unseen classes is to transfer knowledge from seen classes, utilizing semantic information. However, existing semantic information suffers from incomplete semantics, which leads to severe domain shaft problems between real and synthetic data. In this paper, we propose a novel two-phase framework to fundamentally augment generative ZSL frameworks via the combination of multimodal semantics. Specifically, we design an unsupervised algorithm to effectively fuse two kinds of semantics: human-defined attributes (HA) and commonsense embeddings (CSE) from ConceptNet. To further fine-tune the accuracy, we apply a cascade classification network consisting of two prototype classifiers to leverage the unseen samples with high confidence. Extensive experiments show that our method can outperform state-of-the-art methods on multiple datasets with significant improvements, and can be a plug-in to augment generative ZSL frameworks.

Recently, convolutional neural networks have yielded promising results in infrared small target detection. However, limited data is a main restriction to the further promotion of detection performance. To solve this issue, we propose a novel two-stage synthetic data augmentation method, involving StyleGAN-based background generation and Transformer-based target fusion, which aims at generating diverse infrared small target images fitting the original distribution. In the background generation stage, we devise a spatial and low-frequency StyleGAN to ameliorate background generation quality, effectively adapting to less high-frequency information in infrared images. In the target fusion stage, a target prior-based Transformer model with a new detecting difficulty distribution similarity loss is proposed to modulate the intensity of targets implanted on synthetic backgrounds. Experimental results show that our synthetic data augmentation method greatly improves the performance of four detection models on three public datasets and attains state-of-the-art results compared to existing data augmentation methods.

Telecommunications fraud incidents have occurred frequently around the world. Rapid and accurate detection of fraudsters in telecommunication networks has become a hot topic for researchers. Due to data imbalance and dependency on labels, many Graph Neural Networks (GNNs)-based methods cannot achieve good representation ability in Telecommunications Fraud Detection (Telecom-FD) tasks. Recently, Graph Contrast Learning (GCL) has emerged as a successful method for fraud detection due to its unsupervised graph representation learning ability. However, existing GCL-based methods typically adopt uniform strategies to drop edges and impose perturbations on features during the data augmentation process, which tends to eliminate certain influential edges, consequently reducing the quality of the embeddings. To overcome the above problems, a novel framework named TFD-GCL is proposed to adaptively generate contrastive instances for effective Telecom-FD. Specifically, encoding anchor nodes and generating positive instances by automatically learning from neighbors using the attention mechanism. Then, the end-to-end adaptive learning global average similarity is devised to evaluate interaction information between neighbors and construct negative instances from dissimilar neighbors. In addition, TFD-GCL uses a joint contrastive loss to improve the representational capability of fraudulent nodes in unbalanced data. Extensive experiments on the real-world dataset demonstrate that TFD-GCL has better performance compared with classical Telecom-FD methods when the labels are extremely limited.

Accelerometer signals play a critical role in many fields, for example navigation and vehicle safety. However, uncontrollable factors such as defective equipment and harsh environments make the signals recorded by sensors contain a large amount of noise, which poses a great challenge. Most existing methods are based on traditional signal processing and often face problems of incomplete noise reduction or signal distortion after denoising. In this paper, we propose a data-driven denoising method for accelerometer signals (ADNet) based on the Wave_U_Net network architecture, incorporating Multi-Head Attention mechanism and Spatial Attention mechanism. This enhances the network’s feature extraction capability and accelerates its convergence speed. Meanwhile, the attention mechanism focuses the model’s attention on the clean signal’s feature information, improving the network’s fitting capability to the signal distribution. Therefore, ADNet can achieve more thorough denoising while reducing signal distortion after denoising. Finally, we conduct experiments on Walking speed dataset and field experiment dataset to verify the effectiveness of ADNet. The experimental results show that ADNet outperforms other baseline models in terms of Mean Square Error, Mean Absolute Error, Root Mean Square Error, and Signal-to-Noise Ratio.

Abstract Efficient animal detection is essential for biodiversity protection. Unmanned aerial vehicles (UAVs) have been widely used because of their low costs and minimal environmental intrusion. However, using UAVs for practical animal detection poses two challenges: (a) the UAV's fly highly to avoid disturbing animals, resulting in small object detection problems; (b) the limited processing power of UAVs makes large state‐of‐the‐art (SOTA) methods (e.g., You Only Look Once V7, YOLOv7) difficult to deploy. This work proposes the WILD‐YOLO based on YOLOv7 to deal with the two problems. To detect small objects, WILD‐YOLO improves upon YOLOv7 by adding a small object detection head in the head part. To enable real‐time animal detection in field environments with UAVs, the lighten FasterNet and GhostNet have been used to significantly reduce the model size. Compared to YOLOv7, WILD‐YOLO significantly reduces the number of parameters, making it suitable for lightweight deployment on UAVs. Additionally, comparisons with other lightweight models such as YOLOv7‐tiny, YOLOv5‐s, YOLOv4‐s and MobilenetV2 on the datasets are conducted. The experimental results demonstrate that this proposed WILD‐YOLO method outperforms other approaches and has great potential for effective detection of wildlife in complex environments encountered by UAVs.

The big data of Internet of Vehicles contributes to the development of intelligent transportation. Privacy protection in vehicular ad hoc networks (VANETs) is the core factor to improve user and vehicle participation. This article proposes a novel blockchain-based dynamic extensible privacy protection and message authentication scheme for VANETs. It minimizes the computation cost of message authentication based on an elliptic curve and message batch verification. Based on the Chinese remainder theorem, this scheme protects transmitted message security by adaptively and dynamically responding to vehicles and roadside units accessing the VANET. It offers a smart contract-based forensics and tracing solution from the accident vehicle. In addition, strict security proof and analysis that the scheme meets the security requirements for the VANET. It evaluates the efficiency of the scheme, and the results show its practicality.

The Internet of Medical Things (IoMT) plays a crucial role in advancing smart healthcare by facilitating the real-time collection and processing of medical data. These interconnected devices leverage Artificial Intelligence to assist practitioners in making data-driven decisions. However, IoMT’s dependence on communication protocols exposes it to significant security vulnerabilities. In response to this challenge, we propose a novel Meta-Intrusion Detection System (Meta-IDS) that employs a meta-learning approach to enhance the detection of both known and zero-day intrusions. Our approach seamlessly integrates signature-based and anomaly-based detection techniques, incorporating privacy-preserving methods essential for handling sensitive IoMT data. We rigorously evaluated our methodology using three publicly available datasets (WUSTL-EHMS-2020, IoTID20, and WUSTL-IIOT-2021). The results demonstrate remarkable accuracy rates of 99.57%, 99.93%, and 99.99% for signature-based detection, and 99.47%, 99.98%, and 99.99% for anomaly-based detection, coupled with impressively low misclassification rates of 0.0042%, 0.0006%, and 0.00004%, respectively. Through a comparative analysis with the state-of-the-art E-GraphSAGE model, considering metrics such as accuracy, precision, recall, F1-score, time complexity, and misclassification rate, we affirm the performance and reliability of the Meta-IDS. Our approach holds significant promise in bolstering cybersecurity within the IoMT network.

In the domain of complex networks, the identification of influential nodes plays a crucial role in ensuring network stability and facilitating efficient information dissemination . Although the study of influential nodes has been applied in many fields such as suppression of rumor spreading, regulation of group behavior , and prediction of mass events evolution, current deep learning-based algorithms have limited input features and are incapable of aggregating neighbor information of nodes, thus failing to adapt to complex networks. We propose an influential node identification method in complex networks based on the Transformer. In this method, the input sequence of a node includes information about the node itself and its neighbors, enabling the model to effectively aggregate node information to identify its influence. Experiments were conducted on 9 synthetic networks and 12 real networks. Using the SIR model and a benchmark method to verify the effectiveness of our approach. The experimental results show that this method can more effectively identify influential nodes in complex networks. In particular, the method improves 27 percent compared to the second place method in network Netscience and 21 percent in network Faa.

Somatic tumors have a high-dimensional, sparse, and small sample size nature, making cancer subtype stratification based on somatic genomic data a challenge. Current methods for improving cancer clustering performance focus on dimension reduction, integrating multi-omics data, or generating realistic samples, yet ignore the associations between mutated genes within the patient-gene matrix. We refer to these associations as gene mutation structural information, which implicitly includes cancer subtype information and can enhance subtype clustering. We introduce a novel method for cancer subtype clustering called SIG(Structural Information within Graph). As cancer is driven by a combination of genes, we establish associations between mutated genes within the same patient sample, pair by pair, and use a graph to represent them. An association between two mutated genes corresponds to an edge in the graph. We then merge these associations among all mutated genes to obtain a structural information graph, which enriches the gene network and improves its relevance to cancer clustering. We integrate the somatic tumor genome with the enriched gene network and propagate it to cluster patients with mutations in similar network regions. Our method achieves superior clustering performance compared to SOTA methods, as demonstrated by clustering experiments on ovarian and LUAD datasets.

Internet of Things (IoT) devices are large in number, widely distributed, weak in protection ability, and vulnerable to various malicious attacks. Intrusion detection technology can provide good protection for network equipment. However, the normal traffic and abnormal traffic in the network are usually imbalanced. Imbalanced samples will seriously affect the performance of machine learning detection algorithm. Therefore, this paper proposes an intrusion detection method based on data augmentation, namely TMG-IDS. We name the proposed data augmentation model TMG-GAN, which is a data augmentation method based on generative adversarial networks (GAN). First, TMG-GAN has a multi-generator structure, which can be used to generate different types of attack data simultaneously. Second, we increase the classifier structure, which can optimize the generator and discriminator more efficiently based on the classification loss. Third, we calculate the cosine similarity between the generated samples and the original samples and other types of generated samples as a generator loss, which can further improve the quality of generated samples and reduce the class overlap area between the distributions of various generated samples. We conduct extensive experiments on two intrusion detection datasets, CICIDS2017 and UNSW-NB15. The experimental results show that compared with the advanced oversampling algorithm and the latest intrusion detection algorithm, the proposed TMG-IDS method has a good detection effect under the three indicators of Precision, Recall and F1-score.

We study the problem of classifying different cooking styles, based on the recipe. The difficulty is that the same food ingredients, seasoning, and the very similar instructions result in different flavors, with different cooking styles. Existing methods have limitations: they mainly focus on homogeneous data (e.g., instruction or image), ignoring heterogeneous data (e.g., flavor compound or ingredient), which certainly hurts the classification performance. This is because collecting enough available heterogeneous data of a recipe is a non-trivial task. In this paper, we present a new heterogeneous data augmentation method to improve classification performance. Specifically, we first construct a heterogeneous recipe graph network to represent heterogeneous data, which includes four main-stream types of heterogeneous data: ingredient, flavor compound, image, and instruction. Then, we draw a sequence of augmented graphs for Semi-Supervised learning through multinomial sampling. The probability distribution of sampling depends on the Cosine distance between the nodes of graph. In this way, we name our approach as Multinomial Sampling Graph Data Augmentation (MS-GDA). Extensive experiments demonstrate that MS-GDA significantly outperforms SOTA baselines on cuisine classification and region prediction with the recipe benchmark dataset. Code is available at https://github.com/LiangzheChen/MS-GDA .

The long-standing one-to-many issue of the open-domain dialogues poses significant challenges for automatic evaluation methods, i.e., there may be multiple suitable responses which differ in semantics for a given conversational context.To tackle this challenge, we propose a novel learning-based automatic evaluation metric (CMN), which can robustly evaluate open-domain dialogues by augmenting Conditional Variational Autoencoders (CVAEs) with a Next Sentence Prediction (NSP) objective and employing Mutual Information (MI) to model the semantic similarity of text in the latent space.Experimental results on two opendomain dialogue datasets demonstrate the superiority of our method compared with a wide range of baselines, especially in handling responses which are distant to the golden reference responses in semantics.

Determining the genres of a trailer is a challenging multi-label classification task. Previous studies tend to classify by CNN or RNN. Recently, Transformer based on attention mechanism has achieved better results in many research fields than CNN and RNN. Inspired by these, we propose a Hierarchical Transformer (HT). HT can process both the frame sequence (HT-F) and audio (HT-A) of trailers. Besides, a feature compression module is inserted into HT-F, and audio spectrogram segment is processed by HT-A as a whole, which can effectively reduce the data processed by the second Transformer. In order to reduce the training cost and improve the performance, we load the pre-trained weights from other related fields into some parameters of HT, and utilize the limited resources to train the remaining parameters. Experiments show that our best model outperforms state-of-the-art methods on several comprehensive metrics.

Imbalanced data cause deep neural networks to output biased results, and it becomes more serious when facing extremely imbalanced data regarding the outliers with tiny size (the ratio of the outlier size to the image size is around 0.05%). Many data argumentation models are proposed to supplement imbalanced data to alleviate biased results. However, the existing augmentation models cannot synthesize tiny outliers, which make the generated data unavailable. In this article, we propose a new augmentation model named extremely imbalanced data augmentation generative adversarial nets (EID-GANs) to address the extremely imbalanced data augmentation problem. First, we design a new penalty function by subtracting the outliers from the cropped region of generated instance to guide the generator to learn the features of outliers. After this, we combine the output value of the penalty function with the generator loss to jointly update the generator’s parameters with backpropagation. Second, we propose a new evaluation approach that adopts two outlier detectors withk-fold cross-validation to assess the availability of generated instances. We conduct extensive experiments to demonstrate the significant performance improvement of EID-GAN on two extremely imbalanced datasets, which are the industrial Piston and the Fabric datasets, and one general imbalanced dataset, i.e., the public DAGM dataset. The experimental results show that our EID-GAN outperforms the state-of-the-art (SOTA) augmentation models on different imbalanced datasets.

The generative adversarial network (GAN) is usually built from the centralized, independent identically distributed (i.i.d.) training data to generate realistic-like instances. In real-world applications, however, the data may be distributed over multiple clients and hard to be gathered due to bandwidth, departmental coordination, or storage concerns. Although existing works, such as federated learning GAN (FL-GAN), adopt different distributed strategies to train GAN models, there are still limitations when data are distributed in a non-i.i.d. manner. These studies suffer from convergence difficulty, producing generated data with low quality. Fortunately, we found that these challenges are often due to the use of a federated averaging strategy to aggregate local GAN models' updates. In this article, we propose an alternative approach to tackling this problem, which learns a globally shared GAN model by aggregating locally trained generators' updates with maximum mean discrepancy (MMD). In this way, we term our approach improved FL-GAN (IFL-GAN). The MMD score helps each local GAN hold different weights, making the global GAN in IFL-GAN getting converged more rapidly than federated averaging. Extensive experiments on MNIST, CIFAR10, and SVHN datasets demonstrate the significant improvement of our IFL-GAN in both achieving the highest inception score and producing high-quality instances.

Automated code generation is a longstanding challenge in both communities of software engineering and artificial intelligence. Currently, some works have started to investigate the functional correctness of code generation, where a code snippet is considered correct if it passes a set of test cases. However, most existing works still model code generation as text generation without considering program-specific information, such as functionally equivalent code snippets and test execution feedback. To address the above limitations, this paper proposes a method combining program analysis with deep learning for neural code generation, where functionally equivalent code snippets and test execution feedback will be considered at the training stage. Concretely, we firstly design several code transformation heuristics to produce different variants of the code snippet satisfying the same functionality. In addition, we employ the test execution feedback and design a test-driven discriminative task to train a novel discriminator, aiming to let the model distinguish whether the generated code is correct or not. The preliminary results on a newly published dataset demonstrate the effectiveness of our proposed framework for code generation. Particularly, in terms of the [email protected] metric, we achieve 8.81 and 11.53 gains compared with CodeGPT and CodeT5, respectively.

As the second stage of the rumor defeat task pipeline, rumor tracking aims to filter tweets related to a specific event. However, due to the different levels of attention of different events, events related to fewer tweets may be masked by related to more tweets. In some research, the researchers give up detecting the events containing a small number of tweets for better results. To this end, we propose a Simple Framework for Contrastive Learning of Rumor Tracking (SimCLRT)-a novel rumor tracking framework that uses contrastive learning to alleviate the cover between tweets. SimCLRT contains three variants SimCLRT-CNN, SimCLRT-Linear, and SimCLRT-RNN. We conduct experiments on the two commonly used rumor tracking datasets PHEME and RumorEval. The results show that SimCLRT completely defeated baselines. Like the detection performance on the events that contain many tweets, SimCLRT can also effectively detect events containing a small number of tweets. Furthermore, we compare and analyze the performance of SimCLRT variants. SimCLRT-CNN is the model that performs best in our experiments. Although SimCLRT-Linear has a slight advantage on the RumorEval dataset, its robustness is weaker than SimCLRT-RNN and SimCLRT-CNN. If in a long text environment, we consider SimCLRT-RNN will perform more competitively.

Data usually resides on a manifold, and the minimal dimension of such a manifold is called its intrinsic dimension. This fundamental data property is not considered in the generative adversarial network (GAN) model along with its its variants; such that original data and generated data often hold different intrinsic dimensions. The different intrinsic dimensions of both generated and original data may cause generated data distribution to not match original data distribution completely, and it certainly will hurt the quality of generated data. In this study, we first show that GAN is often unable to generate simulation data, holding the same intrinsic dimension as the original data with both theoretical analysis and experimental illustration. Next, we propose a new model, called Hausdorff GAN, which removes the issue of different intrinsic dimensions and introduces the Hausdorff metric into GAN training to generate higher quality data. This provides new insights into the success of Hausdorff GAN. Specifically, we utilize a mapping function to map both original and generated data into the same manifold. We then calculate the Hausdorff distance to measure the difference between the mapped original data and the mapped generated data, toward pushing generated data to the side of original data. Finally, we conduct extensive experiments (using MNIST, CIFAR10, and CelebA datasets) to demonstrate the significant performance improvement of the Hausdorff GAN in achieving the largest Inception Score and the smallest Frechet inception distance (FID) score as well as producing diverse generated data at different resolutions.

Manipulating facial images results in negative influences on the social association, with deep generative models. Although many detection methods have been proposed, they have either designed sophisticated neural networks that lack enough interpretability, or found defects specific to one manipulation method. To address this issue, we propose a new approach to explore the defects of fake facial images after wavelet transform and call it GFS (Grayscale Frequency Statistics). First, we utilize Haar wavelet transformation to decompose the image into low-frequency approximation, horizontal detail, vertical detail, and diagonal detail. The GFS of real and fake images exhibit different distribution and forms in these four subbands. We qualitatively analyze these differences and quantify them as weights. Then, these four subband images are used to train four CNNs respectively, and the obtained detection results also verify the differences in GFS. After that, we combine the prediction results of the four CNNs and the corresponding weights to further improve the detection performance. We conduct extensive experiments on 11 datasets generated by various facial manipulation methods, and the superior results show the effectiveness of our proposed approach. Our findings indicate that the fake images generated by the current facial manipulation methods cannot simulate real images in wavelet-domain.

The huge amount of information on social media contains much false information that has not been confirmed or has been confirmed but not known by all users, which may cause improper public attention or mislead the lives of Internet users. Among the widely spread false information, there are many rumors that require professional knowledge to be distinguished. In this paper, through in-depth analysis of the data collected from Sina Weibo official false news refuting account and questionnaires sent out by us, we propose the indicator of the relative influence of scientific rumors, which demonstrates that the rumors which requires professional scientific knowledge for identification are small but have strong propagation ability. This makes it particularly important to use technical manners to automatically detect them from a great amount of information. Targeting at this, we build up an Internet rumor dataset which contains three categories: the first is called scientific rumors which needs to be clarified by professional knowledge; the second is named as social rumors that need to be validated by authoritative media, organizations, or officials; and the third is ordinary texts. A long short term memory (LSTM)-based model is developed to detect scientific rumors and social rumors among the collected data. A benchmark on the proposed dataset is provided by comparing multiple competitors, which shows that our model is competitive to the state-of-the-arts. As far as we know, this is the first work that detects scientific rumors uses deep learning tools.

The barriers of food enterprises and departments caused information asymmetry, which is the root cause of food safety incidents. Simultaneously, it is challenging to solve the information asymmetry by the existing cloud-based food supply-chain regulation system. Establishing a secure and reliable data sharing environment is an effective solution to the information island. Blockchain can construct a security network based on mathematical algorithms, eliminating the third party’s potential security risk, and realize transparently share data. In this paper, on the principle of metadata remaining in the food enterprises, we propose a blockchain-cloud fusion scheme based on Decentralized Attribute-Based Signature (DABS) to realize secure data sharing between departments. It constructs a decentralized and trusting environment for data owners to share data and achieves social co-governance of food safety based on the smart contract. It can also preserve the existing system architecture and complement the performance disadvantage of blockchain and cloud storage. The result achieved from security analysis shows that our scheme supports unconditional full anonymity and can resist collusion attacks of N-1 out of N corrupted attribute authorities.

Pooling is an important technique for learning text representations in many neural NLP models. In conventional pooling methods such as average, max and attentive pooling, text representations are weighted summations of the L1 or L∞ norm of input features. However, their pooling norms are always fixed and may not be optimal for learning accurate text representations in different tasks. In addition, in many popular pooling methods such as max and attentive pooling some features may be over-emphasized, while other useful ones are not fully exploited. In this paper, we propose an Attentive Pooling with Learnable Norms (APLN) approach for text representation. Different from existing pooling methods that use a fixed pooling norm, we propose to learn the norm in an end-to-end manner to automatically find the optimal ones for text representation in different tasks. In addition, we propose two methods to ensure the numerical stability of the model training. The first one is scale limiting, which re-scales the input to ensure non-negativity and alleviate the risk of exponential explosion. The second one is re-formulation, which decomposes the exponent operation to avoid computing the real-valued powers of the input and further accelerate the pooling operation. Experimental results on four benchmark datasets show that our approach can effectively improve the performance of attentive pooling.

Automated service classification is the foundation for service discovery and service composition. Currently, many existing methods extracting features from functional description documents suffer the problem of data sparsity. However, beside functional description documents, the Web API ecosystem has accumulated a wealth of information that can be used to improve the accuracy of Web service (API) classification. At the moment, there is an absence of a unified way to combine functional description documents with other sources of information (e.g., attributes, interactions and external knowledge) accumulated in the Web API ecosystem for API classification. To address this issue, we present a dual-GCN framework that can effectively suppress the noise propagation of textual contents by distinguishing functional description documents and other sources of information (specifically Mashup-API co-invocation patterns by default in this paper) for API classification. This framework is extensible with the ability to include different sources of information accumulated in the Web API ecosystem. Comprehensive experiments on a real-world public dataset demonstrate that our proposed method can outperform various representative methods for API classification.

In recent decades, radio frequency identification (RFID) has been used in many applications in the world. Tag anti-collision is a fundamental technique for RFID, for solving the collisions when multiple RFID tags transmit their IDs to an RFID reader simultaneously. This technique is well investigated in the stationary environment, however, has some deficiencies when the number of tags in the interrogation region of the reader changes dramatically. This paper proposes a tag anti-collision algorithm called TAD to solve this problem. TAD can effectively and fast estimate the number of arriving and leaving tags, and automatically adapt to different changes of the tags using a hybrid method. The simulation results show that TAD significantly outperforms existing approaches in the situation with many leaving tags.

The purpose of this special issue is to collate a selection of representative research articles that were primarily presented at the Intelligent Systems for the Internet of Things, held in Wuhan. This annual workshop brings together practitioners, researchers, vendors, and scholars from the complementary fields of IoT, in order to promote an exchange of ideas, discuss future collaborations, and develop new research directions. Many other technologies, such as data mining, big data analytics, statistical, and other analysis techniques, have also been used for exploring data generated from the IoT. In addition to analysis techniques, intelligent technologies, such as simulated annealing, tabu search, genetic algorithm, ant colony optimization, and particle swarm optimization, also provide many possibilities for the IoT because not only can they be used to enhance the performance of a system and its devices but they can also be used to make a system aware of events occurred. This special issue presents many examples of how researchers, scholars, vendors, and practitioners are collaborating to address high performance computing research challenges. The scope of this special issue is broad and is representative of the multi-disciplinary nature of IoT. In addition to submissions that deal with intelligent technologies for the IoT and their applications, this issue also includes articles that address practical challenges with IoT-related technologies, such as cloud computing, wireless sensor network, and big data technologies. In the article Three-dimensional sound reproduction in vehicle based on data mining technique,1 sound reproduction method in vehicle is well researched by engineers. Both the immersive and envelope perception are vital cues in vehicle listening environment. Motor companies such as Cadillac, BMW, FORD, Honda, HYUNDAI develop dozens of loudspeaker systems. Loudspeaker corporations such as Sony, Dolby, Harman/Kardon, bose manufacture lots of high-quality loudspeakers for vehicle. However, there are few 3D sound systems in vehicle. Unfortunately, the state-of-the-art 3D sound reproduction techniques are not suitable for vehicle. This topic develops a novel 3D sound reproduction method based on stepwise regression model to generate 3D sounds. The experiments show the validation of the proposed method. The authors provide an alternative 3D sound reproduction technique for both vehicle and loudspeaker manufactures. In the article Loudspeaker triplet selection based on low distortion within head for multichannel conversion of smart 3D home theater,2 the key to the Smart 3D home theater is that you can experience the 3D audio and visual effects of the theater at home. However, in the theater, the popular 3D multichannel sound systems (such as Dolby Atoms, Barco Auro-3D, NHK 22.2 multichannel sound system, etc) cannot be converted directly into a multichannel audio system of the Smart 3D home theater while maintaining the 3D audient experience. The author introduced an optimal loudspeaker triplet selection that maintained the theater's 3D sound immersive experience while converting the theater's 3D multichannel sound system into a home multichannel sound system. Experiments have shown that the distortion of the reproduced sound field within the listener area will be effectively controlled by the proposed method. Therefore, the proposed method is superior to other similar methods in the 3D multichannel sound system conversion. In the article A new source-filter model audio bandwidth extension using high frequency perception feature for IoT communications,3 audio coding is a generic technology to IoT applications system. Audio bandwidth extension (ABE) is a standard technique within contemporary audio codecs to efficiently code audio signals at low bitrates. In traditional methods, only the current frame low-frequency signal is used to generate high-frequency signal, which will result in the perception coding quality degrade significantly if the correlation between low and high frequency becomes weak. Jiang et al proposed a new source-filter model ABE method. In their method, a perception feature of the high-frequency signal is extracted to restore the perception quality of coding. In the ABE decoder side, the crest factor and noise level are obtained under the constraint of the high-frequency perception parameter. Their experiment results shown are superior to the traditional methods and are also comparative compared with the state-of-the-art methods. This paper provided a new solution for IoT communications that requires low bitrates and high quality of audio coding. In the article Users' location analysis based on Chinese mobile social media,4 Wang et al have revised UGC-LI model in preprocess of texts and the creation of the local dictionaries that take existed local dictionaries from the Internet into consideration, with the purpose of the inferences for users and texts locations. Through the crawler, the model acquires users' personal information, the blog content, and customer relationships (follows, fans) information data more than 410 331 pieces from Sina Weibo. The experimental results show that the recall rate of the user location inference is 86.0%, whereas the precise rate is 77.4%, and the accuracy of text posted location inference is 66.8%. Compared with some other related algorithms, this revised model has comparatively better results in location inference for users and texts publication.

Summary After the rapid development for more than 20 years, Internet has gradually become the main carrier of people's information and behaviors in people's daily life. In addition, the innovation and popularization of smartphone GPS makes user location information much more available and accurate, helping it to create remarkable values by which people are attracted to focus on social media–related data mining and applications. However, because of the sparsity of social media geographical information, direct inferences of locations have plenty of difficulties. Under the background of big data, this research has revised the UGC‐LI model in the preprocess of texts and the creation of the local dictionaries in which we take existed local dictionaries from the Internet into consideration, with the purpose of the inferences for users' and texts' locations. At the time of writing, through the crawler, we acquire users' personal information, the blog content, and customer relationships' (follows, fans) information more than 410 331 pieces from Sina Weibo. The experimental results show that the recall rate of the user location inference is 86.0%, whereas the precise rate is 77.4%, and the accuracy of text posted location inference is 66.8%. Compared with some other related algorithms, this revised model has comparatively better results in location inference for users and text publication.

Context: Ranking-oriented defect prediction (RODP) ranks software modules to allocate limited testing resources to each module according to the predicted number of defects. Most RODP methods overlook that ranking a module with more defects incorrectly makes it difficult to successfully find all of the defects in the module due to fewer testing resources being allocated to the module, which results in much higher costs than incorrectly ranking the modules with fewer defects, and the numbers of defects in software modules are highly imbalanced in defective software datasets. Cost-sensitive learning is an effective technique in handling the cost issue and data imbalance problem for software defect prediction. However, the effectiveness of cost-sensitive learning has not been investigated in RODP models. Aims: In this article, we propose a cost-sensitive ranking support vector machine (SVM) (CSRankSVM) algorithm to improve the performance of RODP models. Method: CSRankSVM modifies the loss function of the ranking SVM algorithm by adding two penalty parameters to address both the cost issue and the data imbalance problem. Additionally, the loss function of the CSRankSVM is optimized using a genetic algorithm. Results: The experimental results for 11 project datasets with 41 releases show that CSRankSVM achieves 1.12%-15.68% higher average fault percentile average (FPA) values than the five existing RODP methods (i.e., decision tree regression, linear regression, Bayesian ridge regression, ranking SVM, and learning-to-rank (LTR)) and 1.08%-15.74% higher average FPA values than the four data imbalance learning methods (i.e., random undersampling and a synthetic minority oversampling technique; two data resampling methods; RankBoost, an ensemble learning method; IRSVM, a CSRankSVM method for information retrieval). Conclusion: CSRankSVM is capable of handling the cost issue and data imbalance problem in RODP methods and achieves better performance. Therefore, CSRankSVM is recommended as an effective method for RODP.

Provides an abstract of the keynote presentation and may include a brief professional biography of the presenter. The complete presentation was not made available for publication as part of the conference proceedings.

Analyzing the crash reports recorded upon software crashes is a critical activity for software quality assurance. Predicting whether or not the fault causing the crash (crashing fault for short) resides in the stack traces of crash reports can speed-up the program debugging process and determine the priority of the debugging efforts. Previous work mostly collected label information from bug-fixing logs, and extracted crash features from stack traces and source code to train classification models for the Identification of Crashing Fault Residence (ICFR) of newly-submitted crashes. However, labeled data are not always fully available in real applications. Hence the classifier training is not always feasible. In this work, we make the first attempt to develop a cross project ICFR model to address the data scarcity problem. This is achieved by transferring the knowledge from external projects to the current project via utilizing a state-of-the-art Balanced Distribution Adaptation (BDA) based transfer learning method. BDA not only combines both marginal distribution and conditional distribution across projects but also assigns adaptive weights to the two distributions for better adjusting specific cross project pair. The experiments on 7 software projects show that BDA is superior to 9 baseline methods in terms of 6 indicators overall.

As a typical task of time series mining, Time Series Classification (TSC) has attracted lots of attention from both researchers and domain experts due to its broad applications. To get rid of costly hand-crafting feature engineering process, deep learning techniques are applied for automatic feature extraction, which shows competitive or even better performance compared with state-of-the-art TSC solutions. However, on time series datasets presenting complex patterns, neither 1-Nearest-Neighbour classifier nor deep learning models are capable of achieving satisfactory classification accuracy which motivates us to explore new time series representations to help classifiers further improve the classification accuracy. In this paper, by building the binary distribution tree, an approach to time series classification based on deep learning models using new representations is proposed. By conducting comprehensive experiments over 6 most challenging time series datasets and comparing experimental results of the same classifier using the proposed representation or not, the potential of the proposed approach to enhancing time series classification accuracy is validated with a bunch of helpful findings.

Summary As a well‐known field of big data applications, smart city takes advantage of massive data analysis to achieve efficient management and sustainable development in the current worldwide urbanization process. An important problem in smart city is how to discover frequent trajectory sequence pattern and cluster trajectory. To solve this problem, this paper proposes a cloud‐based taxi trajectory pattern mining and trajectory clustering framework for smart city. Our work mainly includes (1) preprocessing raw Global Positioning System trace by calling the Baidu API Geocoding; (2) proposing a distributed trajectory pattern mining (DTPM) algorithm based onSpark; and (3) proposing a distributed trajectory clustering (DTC) algorithm based onSpark. The proposed DTPM algorithm and DTC algorithm can overcome the high input/output overhead and communication overhead by adopting in‐memory computation. In addition, the proposed DTPM algorithm can avoid generating redundant local trajectory patterns to significantly improve the overall performance. The proposed DTC algorithm can enhance the performance of trajectory similarity computation by transforming the trajectory similarity calculation into AND and OR operators. Experimental results indicate that DTPM algorithm and DTC algorithm can significantly improve the overall performance and scalability of trajectory pattern mining and trajectory clustering on massive taxi trace data. Copyright © 2016 John Wiley & Sons, Ltd.

The vertical profile of atmospheric CO2is of great scientific significance in identifying carbon sinks and sources, and estimating CO2emissions or uptakes. Differential absorption Light Detection And Ranging (DIAL), has been widely accepted as the most promising technique to sense atmospheric CO2. The classical method to retrieve measurements, generated from range-resolved detection, is derived from differentiating the measured column content, but its performance in dealing with aerosol backscatter signals is poor. To address this issue, this paper proposes a derivative method, which is based on Chebyshev fitting to the measured differential absorption optical depth. We created a performance evaluation model to assess the performance of the proposed method. Simulations revealed that the error of a single CO2profile in data retrieval can be reduced to less than 4 ppm in 6000 m. The precision of long-term mean CO2profile is expected to be less than 1 ppm. We believe that this novel method can be used in other applications also, e.g., trace gas measurements collected using DIAL, especially when the signal-to-noise-ratio of received signal is small.

The vertical profile of atmospheric carbon dioxide is one of the most significant parameters for carbon cycle study. Ground-based differential absorption lidar (DIAL) is widely accepted as the most compromising technique to obtain vertical profile of atmospheric CO2. A container-integrated prototype was demonstrated in this paper. Both the configuration and preliminary results were presented here. This equipment will be tested along with the other 6 equipments of atmospheric soundings in Huainan this year, aiming to deepen our understanding about the interactions within neutral atmosphere layers.

Defect prediction aims to estimate software reliability via learning from historical defect data. A defect prediction method identifies whether a software module is defect-prone or not according to metrics that are mined from software projects. These metric values, also known as features, may involve irrelevance and redundancy, which will hurt the performance of defect prediction methods. Existing work employs feature selection to preprocess defect data to filter out useless features. In this paper, we propose a novel feature selection framework, MICHAC, short for defect prediction via Maximal Information Coefficient with Hierarchical Agglomerative Clustering. MICHAC consists of two major stages. First, MICHAC employs maximal information coefficient to rank candidate features to filter out irrelevant ones, second, MICHAC groups features with hierarchical agglomerative clustering and selects one feature from each resulted group to remove redundant features. We evaluate our proposed method on 11 widelystudied NASA projects and four open-source AEEEM projects using three different classifiers with four performance metrics (precision, recall, F-measure, and AUC). Comparison with five existing methods demonstrates that MICHAC is effective in selecting features in defect prediction.

Service computing is a popular development paradigm in information technology. The functional properties of Web services assure correct functionality of cloud applications, while the nonfunctional properties such as reliability might significantly influence the user-perceived availability evaluation. Reliability rankings provide valuable information for making optimal cloud service selection from a set of functionally-equivalent candidate services. There existed several approaches that can conduct reliability ranking prediction for Web services. Those approaches acquire different rankings with different preference functions. It is arduous to determine whether there exists the best one in them, and what is the best one if not. This paper proposes a learning approach to reliability ranking prediction for Web services which utilizes past service invocation logs to train preference function. To validate the proposed approach, large-scale experiments are conducted based on a real-world Web service dataset, WSDream. The results show that our proposed approach achieves higher prediction accuracy than the existing approaches.

Wireless sensor networks (WSNs) are vulnerable to the unfavorable funneling effect. The optimization of WSN clustering is a natural way to suppress the funneling effect. WSN clusters involve the edge effect that was undervalued in existing techniques. We propose an optimal clustering routing protocol GreenOCR to reduce the detrimental influence of the funnel effect and minimize the energy consumption in WSNs. Our work focuses on the approximate unequal optimal clustering and dropping energy consumption arising from the edge effect. First, according to the data repeat rate among overlapped clusters, we estimate the actual data compression ratio to offset the negative influence of the edge effect and save WSN energy. Secondly, we reduce the issue of minimizing the total energy consumption in a WSN to a nonlinear programming (NLP). We have proved that this NLP problem is NP complete. Third, we turn over to exploring an approximate optimal clustering and propose an approximate optimal clustering algorithm. A GreenOCR enabled WSN clustering minimizes the energy consumption in the whole network and extends the lifetime of the WSN. The simulation experiment shows that GreenOCR outperforms its rivals in alleviating the funnel effect.

Summary The precise prediction of bus routes or the arrival time of buses for a traveler can enhance the quality of bus service. However, many social factors influence people's preferences for taking buses. These social factors may include heavy traffic cost, traffic congestion, poor air quality and so forth. Existing prediction techniques rarely consider social sensing when predicting the bus arrival time. Accordingly, this paper proposes a social sensing enhanced service for predicting bus routes, which integrates sensing ability and social networks to understand and measure the influence between social events and vehicle velocity. We focus on the analysis of two different attributions: PT service quality attributions PEAs and road condition attributions PRCAs. Both of them synthesize the social sensing in their evaluation of bus routes. PEA represents individual preferences and PRCA represents physical factors that significantly influence vehicle velocity. Bus relevant social events were further categorized into PEA events or PRCA events. PEAs of buses were scored according to the tendency of bus conditions reflected in social events. Furthermore, an artificial neural network prediction model is established to estimate the bus travel time. Copyright © 2015 John Wiley & Sons, Ltd.

Recent years have witnessed a rapid growth in using Web services for data publishing and sharing among organizations. To improve the efficiency of software development and economize on human and material resources, service reuse is viewed as a powerful means which will not only reuse atomic services, but also reuse arbitrary granularities of Service Process Fragments (SPFs). However, effectively reusing arbitrary granularities of SPFs has not been solved yet. In this paper, we propose a novel method of SPF reuse, named SCKY, based on the Cocke-Kasami-Younger (CKY) algorithm. We first present an extended CKY to do SPF-query. Then we address how to do SPF-query by a probability CKY, i.e. return a SPF with maximum emergence probability. Finally, we explore the QoS Query of SPF. Through a set of experiments, the effectiveness and robustness of our approach are evaluated, where the dataset is constructed by the Web Service Challenge Testset Generator1 (CTG).

With the advance of RFID technology and pervasive computing, a growing number of RFID devices are deployed in the surrounding environment and form large-scale RFID systems. Many applications run on top of such a system, and perform diverse and possibly conflicting data collection tasks. Existing works about RFID data collection either focus on deducing events of interest from primitive data, or scheduling the activation of readers to mitigate various of interference. The former ones assume that the primitive data have been collected already, and the later ones assume that all the readers belong to a single application whose objective is to read all the tags once. It lacks an effective way to specify the constraints in the process of data collection for multiple applications, and coordinate the readers to meet such requirements. In this paper, we proposed a specification language and a reader coordination algorithm to solve this problem. Our language can be used to specify complex constraints in data collection tasks, based on attribute selection, set relations, and temporal relations. And then a permission based data collection approach is developed for the readers to meet these constraints in a distributed way. Extensive simulation results show that the proposed approach outperforms existing approaches in terms of the execution time.

In this paper, a bottom–up vehicle emission model is proposed to estimate real-time$\hbox{CO}_{2}$emissions using intelligent transportation system (ITS) technologies. In the proposed model, traffic data that were collected by ITS are fully utilized to estimate detailed vehicle technology data (e.g., vehicle type) and driving pattern data (e.g., speed, acceleration, and road slope) in the road network. The road network is divided into a set of small road segments to consider the effects of heterogeneous speeds within a road link. A real-world case study in Beijing, China, is carried out to demonstrate the applicability of the proposed model. The spatiotemporal distributions of$ \hbox{CO}_{2}$emissions in Beijing are analyzed and discussed. The results of the case study indicate that ITS technologies can be a useful tool for real-time estimations of$\hbox{CO}_{2}$emissions with a high spatiotemporal resolution.

Detecting and identifying security events to provide cyber situation awareness has become an increasingly important task within the network research and development community. We propose a graph similarity-based approach to event detection and identification that integrates a number of techniques to collect time-varying situation information, extract correlations between event attributes, and characterize and identify security events. Diverging from the traditional rule- or statistical-based pattern matching techniques, the proposed mechanism represents security events in a graphical form of correlation networks and identifies security events through the computation of graph similarity measurements to eliminate the need for constructing user or system profiles. These technical components take fundamentally different approaches from traditional empirical or statistical methods and are designed based on rigorous computational analysis with mathematically proven performance guarantee. The performance superiority of the proposed mechanism is demonstrated by extensive simulation and experimental results.

Document clustering plays an important role in data mining systems. Recently, a flocking-based document clustering algorithm has been proposed to solve the problem through simulation resembling the flocking behavior of birds in nature. This method is superior to other clustering algorithms, including k-means, in the sense that the outcome is not sensitive to the initial state. One limitation of this approach is that the algorithmic complexity is inherently quadratic in the number of documents. As a result, execution time becomes a bottleneck with large number of documents. In this paper, we assess the benefits of exploiting the computational power of Beowulf-like clusters equipped with contemporary Graphics Processing Units (GPUs) as a means to significantly reduce the runtime of flocking-based document clustering. Our framework scales up to over one million documents processed simultaneously in a sixteen-node moderate GPU cluster. Results are also compared to a four-node cluster with higher-end GPUs. On these clusters, we observe 30X-50X speedups, which demonstrate the potential of GPU clusters to efficiently solve massive data mining problems. Such speedups combined with the scalability potential and accelerator-based parallelization are unique in the domain of document-based data mining, to the best of our knowledge.

Abstract Ad hoc networks are very helpful in situations when no fixed network infrastructure is available, such as natural disasters and military conflicts. In such a network, all wireless nodes are equal peers simultaneously serving as both senders and routers for other nodes. Therefore, how to route packets through reliable paths becomes a fundamental problems when behaviors of certain nodes deviate from wireless ad hoc routing protocols. We proposed a novel Dirichlet reputation model based on Bayesian inference theory which evaluates reliability of each node in terms of packet delivery. Our system offers a way to predict and select a reliable path through combination of first‐hand observation and second‐hand reputation reports. We also proposed moving window mechanism which helps to adjust ours responsiveness of our system to changes of node behaviors. We integrated the Dirichlet reputation into routing protocol of wireless ad hoc networks. Our extensive simulation indicates that our proposed reputation system can improve good throughput of the network and reduce negative impacts caused by misbehaving nodes. Copyright © 2010 John Wiley & Sons, Ltd.

The Particle Swarm Optimization (PSO) clustering algorithm can generate more compact clustering results than the traditional K-means clustering algorithm. However, when clustering high dimensional datasets, the PSO clustering algorithm is notoriously slow because its computation cost increases exponentially with the size of the dataset dimension. Dimensionality reduction techniques offer solutions that both significantly improve the computation time, and yield reasonably accurate clustering results in high dimensional data analysis. In this paper, we introduce research that combines different dimensionality reduction techniques with the PSO clustering algorithm in order to reduce the complexity of high dimensional datasets and speed up the PSO clustering process. We report significant improvements in total runtime. Moreover, the clustering accuracy of the dimensionality reduction PSO clustering algorithm is comparable to the one that uses full dimension space.

Particle swarm optimization (PSO) is a population-based stochastic optimization technique, which can be used to find an optimal, or near optimal, solution to a numerical and qualitative problem. In PSO algorithm, the problem solution emerges from the interactions among many simple individual agents called particles. In the real world, we have to frequently deal with searching and tracking an optimal solution in a dynamical and noisy environment. This demands that the algorithm not only find the optimal solution but also track the trajectory of the non-stationary solution. The traditional PSO algorithm lacks the ability to track the changing optimal solution in a dynamic and noisy environment. In this paper, we present a distributed adaptive PSO (DAPSO) algorithm that can be used to track a non-stationary optimal solution in a dynamically changing and noisy environment.

To better understand insurgent activities and asymmetric warfare, a social adaptive model for modeling multiple insurgent groups attacking multiple military and civilian targets is proposed and investigated. This report presents a pilot study using the particle swarm modeling, a widely used non-linear optimal tool, to model the emergence of insurgency campaign. The objective of this research is to apply the particle swarm metaphor as a model of insurgent social adaptation for the dynamically changing environment and to provide insight and understanding of insurgency warfare. Our results show that unified leadership, strategic planning, and effective communication between insurgent groups are not the necessary requirements for insurgents to efficiently attain their objectives.

Intelligence analysts are currently overwhelmed with the amount of information streams generated everyday. There is a lack of comprehensive tool that can real-time analyze the information streams. Document clustering analysis plays an important role in improving the accuracy of information retrieval. However, most clustering technologies can only be applied for analyzing the static document collection because they normally require a large amount of computation resource and long time to get accurate result. It is very difficult to cluster a dynamic changed text information streams on an individual computer. Our early research has resulted in a dynamic reactive flock clustering algorithm which can continually refine the clustering result and quickly react to the change of document contents. This character makes the algorithm suitable for cluster analyzing dynamic changed document information, such as text information stream. Because of the decentralized character of this algorithm, a distributed approach is a very natural way to increase the clustering speed of the algorithm. In this paper, we present a distributed multi-agent flocking approach for the text information stream clustering and discuss the decentralized architectures and communication schemes for load balance and status information synchronization in this approach.

In the real world, we have to frequently deal with searching for and tracking an optimal solution in a dynamic environment. This demands that the algorithm not only find the optimal solution but also track the trajectory of the solution in a dynamic environment. Particle swarm optimization (PSO) is a population-based stochastic optimization technique, which can find an optimal, or near optimal, solution to a numerical and qualitative problem. However, the traditional PSO algorithm lacks the ability to track the optimal solution in a dynamic environment. In this paper, we present a modified PSO algorithm that can be used for tracking a non-stationary optimal solution in a dynamically changing environment.

Fast and high-quality document clustering algorithms play an important role in effectively navigating, summarizing, and organizing information. Recent studies have shown that partitional clustering algorithms are more suitable for clustering large datasets. However, the K-means algorithm, the most commonly used partitional clustering algorithm, can only generate a local optimal solution. In this paper, we present a particle swarm optimization (PSO) document clustering algorithm. Contrary to the localized searching of the K-means algorithm, the PSO clustering algorithm performs a globalized search in the entire solution space. In the experiments we conducted, we applied the PSO, K-means and hybrid PSO clustering algorithm on four different text document datasets. The number of documents in the datasets ranges from 204 to over 800, and the number of terms ranges from over 5000 to over 7000. The results illustrate that the hybrid PSO algorithm can generate more compact clustering results than the K-means algorithm.

We provide a biasing expansion swarm approach (BESA) for using multiple simple mobile agents, with limited sensing and communication capabilities, to collaboratively search and locate an indeterminate number of emission sources in an unknown large-scale area. The key concept in this approach is swarm behavior. By applying the three properties of the swarm behavior: separation, cohesion and alignment, our approach can ensure the agent group attains dynamically stable ad-hoc connectivity and fast target convergence. Using a grid map to represent the unknown environment, an ad-hoc network for wireless communication and our biasing expansion algorithm for path planning, each agent simultaneously considers all concentration values collected by other swarm members and determines the positive gradient direction of the whole coverage area of the swarm. This will make the swarm immune to the random sensor errors, local aerosol accumulations and other local interference effects during their search. We present a simulated environment that has multiple emission sources and complex aerosol accumulation and distribution. Based on the simulation, our approach can achieve better performance than the gradient descent approach, which currently appears to be the most popular algorithm for emission source localization.

We describe a swarm-based fuzzy logic control (FLC) mobile sensor network approach for collaboratively locating the hazardous contaminants in an unknown large-scale area. The mobile sensor network is composed of a collection of distributed nodes (robots), each of which has limited sensing, intelligence and communication capabilities. An ad-hoc wireless network is established among all nodes, and each node considers other nodes as extended sensors. By gathering other nodes' locations and measurement data, each node's FLC can independently determine its next optimal deployment location. Simultaneously, by applying the three properties of the swarm behavior: separation, cohesion and alignment, the approach can ensure the sensor network attains wide regional coverage and dynamically stable connectivity. The simulation presented in this paper shows the swarm-based FLC mobile sensor network can achieve better performance and have higher fault tolerance in the event of partial node failures and sensor measurement errors.