Unmanned aerial vehicle (UAV)-aided mobile edge computing (MEC) with a blockchain consensus algorithm is a promising solution for addressing mobile offloading of computation-intensive or latency-sensitive tasks with extensive service range, while ensuring the authenticity of the offloading. Existing UAV-aided MEC studies focus on task offloading or trajectory planning of single-antenna UAVs, missing multiple-input multiple-output (MIMO) UAV systems. In addition, most blockchain-based UAV-aided edge computing studies adopt energy-consuming proof-of-work-based consensus mechanisms, which are unsuitable for energy-limited UAV scenarios. To address the above issues, in this paper, we develop a joint deadline-aware task offloading and UAV trajectory planning scheme, utilizing the effects simulated through an energy-efficient consensus protocol. In order to prevent the exhaustion of UAVs' energy and ensure the authenticity of decision-making, we propose an energy-based Raft (E-Raft) consensus algorithm, enabling dynamic leader selection through a series of decreasing energy thresholds. Subsequently, we present a computational profit maximization problem to jointly optimize deadline-aware task offloading and the UAV swarm's trajectory planning. To address this NP-hard problem, we develop an online priority-based task offloading and trajectory selection algorithm, which is performed by the selected leader of the E-Raft consensus in each round. Simulation results demonstrate that our proposed scheme achieves up to 40% performance improvement over other approaches.

Some of the latest released Code Large Language Models (Code LLMs) have been trained on repository-level code data, enabling them to perceive repository structures and utilize cross-file code information. This capability allows us to directly concatenate the content of repository code files in prompts to achieve repository-level code completion. However, in real development scenarios, directly concatenating all code repository files in a prompt can easily exceed the context window of Code LLMs, leading to a significant decline in completion performance. Additionally, overly long prompts can increase completion latency, negatively impacting the user experience. In this study, we conducted extensive experiments, including completion error analysis, topology dependency analysis, and cross-file content analysis, to investigate the factors affecting repository-level code completion. Based on the conclusions drawn from these preliminary experiments, we proposed a strategy called **Hierarchical Context Pruning (HCP)** to construct high-quality completion prompts. We applied the **HCP** to six Code LLMs and evaluated them on the CrossCodeEval dataset. The experimental results showed that, compared to previous methods, the prompts constructed using our **HCP** strategy achieved higher completion accuracy on five out of six Code LLMs. Additionally, the **HCP** managed to keep the prompt length around 8k tokens (whereas the full repository code is approximately 50k tokens), significantly improving completion throughput. Our code and data will be publicly available.

Haonan Zhang, Run Luo, Xiong Liu, Yuchuan Wu, Ting-En Lin, Pengpeng Zeng, Qiang Qu, Feiteng Fang, Min Yang, Lianli Gao, Jingkuan Song, Fei Huang, Yongbin Li. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Recent studies show that graph neural networks (GNNs) are vulnerable to backdoor attacks. Existing backdoor attacks against GNNs use fixed-pattern triggers and lack reasonable trigger constraints, overlooking individual graph characteristics and rendering insufficient evasiveness. To tackle the above issues, we propose ABARC, the firstAdaptiveBackdoorAttack withReasonableConstraints, applying to both graph-level and node-level tasks in GNNs. For graph-level tasks, we propose a subgraph backdoor attack independent of the graph's topology. It dynamically selects trigger nodes for each target graph and modifies node features with constraints based on graph similarity, feature range, and feature type. For node-level tasks, our attack begins with an analysis of node features, followed by selecting and modifying trigger features, which are then constrained by node similarity, feature range, and feature type. Furthermore, an adaptive edge-pruning mechanism is designed to reduce the impact of neighbors on target nodes, ensuring a high attack success rate (ASR). Experimental results show that even with reasonable constraints for attack evasiveness, our attack achieves a high ASR while incurring a marginal clean accuracy drop (CAD). When combined with the state-of-the-art defense randomized smoothing (RS) method, our attack maintains an ASR over 94%, surpassing existing attacks by more than 7%.

Relation extraction (RE) tends to struggle when the supervised training data is few and difficult to be collected. In this article, we elicit relational and factual knowledge from large pretrained language models (PLMs) for few-shot RE (FSRE) with prompting techniques. Concretely, we automatically generate a diverse set of natural language templates and modulate PLM's behavior through these prompts for FSRE. To mitigate the template bias which leads to unstableness of few-shot learning, we propose a simple yet effective template regularization network (TRN) to prevent deep networks from over-fitting uncertain templates and thus stabilize the FSRE models. TRN alleviates the template bias with three mechanisms: 1) an attention mechanism over mini-batch to weight each template; 2) a ranking regularization mechanism to regularize the attention weights and constrain the importance of uncertain templates; and 3) a template calibration module with two calibrating techniques to modify the uncertain templates in the lowest-ranked group. Experimental results on two benchmark datasets (i.e., FewRel and NYT) show that our model has robust superiority over strong competitors. For reproducibility, we will release our code and data upon the publication of this article.

The rapid expansion of the Internet of Things (IoT) introduces critical challenges in scalability, mobility, and security, particularly in large-scale deployments. While information-centric networking (ICN) addresses these by enhancing content mobility, multipath support, and edge-embedded caching with inherent security features, it faces limitations in handling large heterogeneous environments due to its in-network caching and content-based forwarding strategies. Software-defined networking (SDN) complements ICN by employing a centralized controller to intelligently orchestrate content caching and forwarding, yet struggles with the efficient allocation and acquisition of content across expansive IoT systems. In response to these challenges, we propose CIC-SIoT, a novel information-centric SDN (IC-SDN) solution, designed to optimize the ICN-IoT framework. Our solution incorporates specialized algorithms for controllers, consumers, producers, and ICN nodes. These algorithms improve content forwarding decisions by moving beyond the traditional reliance on the forwarding information base (FIB) and instead utilizing the pending interest table (PIT) to efficiently manage and distribute content. Validated through ndnSIM and MATLAB simulations, CIC-SIoT achieves substantial performance enhancements, including an 80% increase in throughput, a 34% reduction in latency, and a 25% savings in bandwidth. Additionally, it reduces packet loss by 67% and communication overhead by 66%, compared to existing solutions. These results underscore the framework’s ability to significantly improve the efficiency and scalability of content distribution in IoT environments, highlighting its robustness and adaptability in addressing the complex dynamics of modern networked systems.

As a promising implementation model of payment channel network (PCN), payment channel hub (PCH) could achieve high throughput by providing stable off-chain transactions through powerful hubs. However, existing PCH schemes assume hubs are preplaced in advance, not considering payment requests' distribution and may affect network scalability, especially network load balancing. In addition, current source routing protocols with PCH allow each sender to make routing decision on his/her own request, which may have a bad effect on performance scalability (e.g., deadlock) for not considering other senders' requests. This paper proposes a novel multi-PCHs solution with high scalability. First, we are the first to study the PCH placement problem and propose optimal/approximation solutions with load balancing for small-scale and large-scale scenarios, by trading off communication costs among participants and turning the original NP-hard problem into a mixed-integer linear programming (MILP) problem solving by supermodular techniques. Then, on global network states and local directly connected clients' requests, a routing protocol is designed for each PCH with a dynamic adjustment strategy on request processing rates, enabling high-performance deadlock-free routing. Extensive experiments show that our work can effectively balance the network load, and improve the performance on throughput by 29.3% on average compared with state-of-the-arts.

Federated learning (FL) protects training data in clients by collaboratively training local machine learning models of clients for a global model, instead of directly feeding the training data to the server. However, existing studies show that FL is vulnerable to various attacks, resulting in training data leakage or interfering with the model training. Specifically, an adversary can analyze local gradients and the global model to infer clients’ data, and poison local gradients to generate an inaccurate global model. It is extremely challenging to guarantee strong privacy protection of training data while ensuring the robustness of model training. None of the existing studies can achieve the goal. In this paper, we propose a robust privacy-preserving federated learning framework (PILE), which protects the privacy of local gradients and global models, while ensuring their correctness by gradient verification where the server verifies the computation process of local gradients. In PILE, we develop a verifiable perturbation scheme that makes confidential local gradients verifiable for gradient verification. In particular, we build two building blocks of zero-knowledge proofs for the gradient verification without revealing both local gradients and global models. We perform rigorous theoretical analysis that proves the security of PILE and evaluate PILE on both passive and active membership inference attacks. The experiment results show that the attack accuracy under PILE is between$[50.3\%,50.9\%]$, which is close to the random guesses. Particularly, compared to prior defenses that incur the accuracy losses ranging from 2% to 13%, the accuracy loss of PILE is negligible, i.e., only$\pm 0.3\%$accuracy loss.

Significant progress has been made by joint entity and relation extraction methods, which directly generate the relation triplets and mitigate the issue of overlapping relations. However, previous models generate the entity-relation triplets solely from input sentences. Such information is insufficient to support the modeling of interactive information between entities and relations. In this paper, we define the features that provide mutual supports for entity and relation detection but can only be accessed at training time as privileged features for relation extraction, and devise two teacher models to exploit privileged entity and relation features, respectively. Meanwhile, we propose a novel contrastive student-teacher learning framework for joint extraction of entities and relations (STER), where a student network is encouraged to amalgamate privileged knowledge from two expert teacher networks that additionally utilize the privileged features, based on contrastive learning. Experiment results on three benchmark datasets (i.e., ADE, SciERC and CoNLL04) demonstrate that STER has robust superiority over competitors and sets state-of-the-art. For reproducibility, we will release the data and source code once the paper is accepted.

How to enable high-performance task offloading and preserve the trust between participants is imperative yet nontrivial to the Cloud-Edge-Device (CED) computing, mainly because the resources are geo-distributed and operated by different parties. Also, the CED participants are highly dynamic and heterogeneous in resource provision and may conflict in interest. This paper proposes BlockChain-empowered CED (BC-CED), a blockchain-empowered collaborative task offloading for CED computing. In BC-CED, blockchain plays a central role in the main functionality of CED, including task offloading, brokerage of resource usage, and incentives. We distinguish the BC-CED from the existing solutions by modifying the blockchain consensus process, enabling the participants to reach an agreement via solving the task offloading problem. For this purpose, we formulate the offloading problem by considering the computation capabilities of candidate nodes and the network performance. BC-CED allows each participant to apply reinforcement learning-based methods to solve this problem and compete for the right of block output by comparing the offloading policy performance and accepting the best policy as the offloading scheme within the next period. We also propose a truthful incentive mechanism to encourage resource contributions in BC-CED and force them to be honest. Extensive tests by implementing our solutions in a commercialized blockchain platform have shown how BC-CED achieves a superior performance in task offloading and blockchain maintenance.

Information technologies (ITs) have been playing an important role in improving our society, and the fast evolution of ITs creates a competitive environment not only for companies but also for regions. Hence, recognizing the future trend of technologies can be effective in decision-making with regard to technology selection and investment. Blockchain technology with its vast and impressive applications has received considerable attention from researchers, investors, and public agencies. The purpose of this research is to investigate blockchain technology to explore its trends according to their classification by use of the World Intellectual Property Organization (WIPO) database. Furthermore, we particularly evaluate the registered patents in the world's most well-known patent databases such as the USA patent database. We drew the current technology trends in blockchain patents by applying the text mining and clustering approach. The results represent that the registered patents in the USA patent database have been achieved in the growth phase. That means, attention to the blockchain is rising nowadays and most patents focused the cryptocurrencies and their applications in finance. However, blockchain technology is in the emergence phase and is evolving by researchers and inventors.

This work takes the lead to study the aspect-level sentiment classificationin the domain adaptation scenario. Given a document of any domains, the model needs to figure out the sentiments with respect to fine-grained aspects in the documents. Two main challenges exist in this problem. One is to build a robust document modeling across domains; the other is to mine the domain-specific aspects and make use of the sentiment lexicon. In this paper, we propose a novel approach Neural Attentive model for cross-domain Aspect-level sentiment CLassification (NAACL), which leverages the benefits of the supervised deep neural network as well as the unsupervised probabilistic generative model to strengthen the representation learning. NAACL jointly learns two tasks: (i) a domain classifier, working on documents in both the source and target domains to recognize the domain information of input texts and transfer knowledge from the source domain to the target domain. In particular, a weakly supervised Latent Dirichlet Allocation model (wsLDA) is proposed to learn the domain-specificaspectandsentiment lexiconrepresentations that are then used to calculate the aspect/lexicon-aware document representations via a multi-view attention mechanism; (ii) an aspect-level sentiment classifier, sharing the document modeling with the domain classifier. It makes use of the domain classification results and the aspect/sentiment-aware document representations to classify the aspect-level sentiment of the document in domain adaptation scenario. NAACL is evaluated on both English and Chinese datasets with the out-of-domain as well as in-domain setups. Quantitatively, the experiments demonstrate that NAACL has robust superiority over the compared methods in terms of classification accuracy and F1 score. The qualitative evaluation also shows that the proposed model is capable of reasonably paying attention to those words that are important to judge the sentiment polarity of the input text given an aspect.

Open-domain dialog generation, which is a crucial component of artificial intelligence, is an essential and challenging problem. In this article, we present a personalized dialog system, which leverages the advantages of multitask learning and reinforcement learning for personalized dialogue generation (MRPDG). Specifically, MRPDG consists of two subtasks: 1) an author profiling module that recognizes user characteristics from the input sentence (auxiliary task) and 2) a personalized dialog generation system that generates informative, grammatical, and coherent responses with reinforcement learning algorithms (primary task). Three kinds of rewards are proposed to generate high-quality conversations. We investigate the effectiveness of three widely used reinforcement learning methods [i.e., Q-learning, policy gradient, and actor-critic (AC) algorithm] in a personalized dialog generation system and demonstrate that the AC algorithm achieves the best results on the underlying framework. Comprehensive experiments are conducted to evaluate the performance of the proposed model on two real-life data sets. Experimental results illustrate that MRPDG is able to produce high-quality personalized dialogs for users with different characteristics. Quantitatively, the proposed model can achieve better performance than the compared methods across different evaluation metrics, such as the human evaluation, BiLingual Evaluation Understudy (BLEU), and perplexity.

Real-time event summarization (RES) aims at extracting a handful of document updates from an overwhelming document stream as the real-time event summary that tracks and summarizes the evolving event of interest. It has been attracting much attention, especially with the growth of streaming applications. Despite the effectiveness of previous studies, obtaining relevant, nonredundant, and timely event summaries remains challenging in real-life applications. This study proposes an effective Hybrid learning model for RES (HRES), which attempts to resolve all three challenges (i.e., nonredundancy, relevance, and timeliness) of RES in a unified framework. The main idea is to: 1) exploit the factual background knowledge from the knowledge base (KB) to capture the informative knowledge and implicit information from the input document/query for better text matching; 2) design a memory network to memorize the input facts temporally from the historical document stream and avoid pushing redundant facts in subsequent timesteps; 3) leverage relevance prediction as an auxiliary task to strengthen the document modeling and help to extract relevant documents; and 4) consider both historical dependencies and future uncertainty of the real-time document stream by exploiting the reinforcement learning technique. Extensive experiments demonstrate that HRES has robust superiority over competitors and gains the state-of-the-art results.

Epidemics especially those caused by major contagious diseases have entailed huge losses in human history. The fights have thus never stopped to prevent pandemics. Due to its acute outbreak, is generally susceptible to the population regardless of ages, so strict quarantine of the infections becomes the most effective means for the epidemic control, which has been proved in the prevention of other contagious diseases such as SARS and H1N1. The key strategy widely used to find infected and suspected patients is still the epidemiological tracking of confirmed cases. However, this may fail to identify infections especially when patients do not show any symptoms. Therefore, the approach to rapid, effective, and simple infection identification is essential to prevent the spread of a contagious disease. This paper proposes to leverage a social apps and Geospatial artificial intelligence (GeoAI) with Blockchain to effectively identify infections with privacy concern. Since people widely use social apps, a large scale of social data with geospatial information could be easily collected and kept on Blockchain with privacy preservation, which thus provides a framework of decentralized, tamper-proof, and privacy-preserved information sharing. With the support of GeoAI, which analyzes the spatial distribution of diseases from the shared data, we could study the influence factors based on spatial propagation of contagious diseases for infection identification. Since WeChat is widely used in China, we take COVID-19 as an example to use the experiments on real-life datasets demonstrate the effectiveness of our method, and provide insight into epidemic control in terms of geo-social data sharing.

Networked data, notably social network data, often comes with a rich set of annotations, or attributes, such as documents (e.g., tweets) and locations (e.g., check-ins). Community search in such attributed networks has been studied intensively due to its many applications in friends recommendation, event organization, advertising, etc. We study the problem of attribute-constrained co-located community (ACOC) search, which returns a community that satisfies three properties: i) structural cohesiveness: the members in the community are densely connected; ii) spatial co-location: the members are close to each other; and iii) attribute constraint: a set of attributes are covered by the attributes associated with the members. The ACOC problem is shown to be NP-hard. We develop four efficient approximation algorithms with guaranteed error bounds in addition to an exact solution that works on relatively small graphs. Extensive experiments conducted with both real and synthetic data offer insight into the efficiency and effectiveness of the proposed methods, showing that they outperform three adapted state-of-the-art algorithms by an order of magnitude. We also find that the approximation algorithms are much faster than the exact solution and yet offer high accuracy.

Question answering (QA) in the legal domain has gained increasing popularity for people to seek legal advice. However, existing QA systems struggle to comprehend the legal context and provide jurisdictionally relevant answers due to the lack of domain expertise. In this paper, we develop an Artificial Intelligence Law Assistant (AILA) for question answering in the domain of Chinese laws. AILA system automatically comprehends users' natural language queries with the help of the legal knowledge graph (KG) and provides the best matching answers for given queries. In addition, AILA provides visual cues to interpret the input queries and candidate answers based on the legal KG. Experimental results on a large-scale legal QA corpus show the effectiveness of AILA. To the best of our knowledge, AILA is the first Chinese legal QA system which integrates the domain knowledge from legal KG to comprehend the questions and answers for ranking QA pairs. AILA is available at http://bmilab.ticp.io:48478/.

Chinese word segmentation (CWS) is an important research topic in information retrieval (IR) and natural language processing (NLP). Significant progresses have been made by deep neural networks with context features. However, these deep models may fail to deal with rare or ambiguous words, thus limit the overall CWS performance. In this paper, we propose a lexicon-enhanced adaptive attention network (LAAN), which takes full advantage of external lexicons to deal with the rare or ambiguous words. Specifically, we devise an adaptive attention mechanism to learn the lexicon-aware representation. In addition, we propose a fusion gate to effectively integrate the additional word information with context information to improve the performance of CWS. LAAN is evaluated on four benchmark datasets, and the experimental results demonstrate that LAAN has robust superiority over the compared methods.

In this paper, we propose a Multi-Task learning approach for Answer Selection (MTAS), motivated by the fact that humans have no difficulty performing such task because they possess capabilities of multiple domains (tasks). Specifically, MTAS consists of two key components: (i) A category classification model that learns rich category-aware document representation; (ii) An answer selection model that provides the matching scores of question-answer pairs. These two tasks work on a shared document encoding layer, and they cooperate to learn a high-quality answer selection system. In addition, a multi-head attention mechanism is proposed to learn important information from different representation subspaces at different positions. We manually annotate the first Chinese question answering dataset in law domain (denoted as LawQA) to evaluate the effectiveness of our model. The experimental results show that our model MTAS consistently outperforms the compared methods.1

In this paper, we propose a Document Embedding Network (DEN) to learn document embeddings in an unsupervised manner. Our model uses the encoder-decoder architecture as its backbone, which tries to reconstruct the input document from an encoded document embedding. Unlike the standard decoder for text reconstruction, we randomly block some words in the input document, and use the incomplete context information and the encoded document embedding to predict the blocked words in the document, inspired by the crossword game. Thus, our decoder can keep the balance between the known and unknown information, and consider both global and partial information when decoding the missing words. We evaluate the learned document embeddings on two tasks: document classification and document retrieval. The experimental results show that our model substantially outperforms the compared methods.1.

The recent artificial intelligence studies have witnessed great interest in abstractive text summarization. Although remarkable progress has been made by deep neural network based methods, generating plausible and high-quality abstractive summaries remains a challenging task. The human-like reading strategy is rarely explored in abstractive text summarization, which however is able to improve the effectiveness of the summarization by considering the process of reading comprehension and logical thinking. Motivated by the humanlike reading strategy that follows a hierarchical routine, we propose a novel Hybrid learning model for Abstractive Text Summarization (HATS). The model consists of three major components, a knowledge-based attention network, a multitask encoder-decoder network, and a generative adversarial network, which are consistent with the different stages of the human-like reading strategy. To verify the effectiveness of HATS, we conduct extensive experiments on two real-life datasets, CNN/Daily Mail and Gigaword datasets. The experimental results demonstrate that HATS achieves impressive results on both datasets.

Various networks have rich attributes such as texts (e.g., tweets) and locations (e.g., check-ins). The community search in such attributed networks have been intensively studied recently due to its wide applications in recommendation, marketing, biology, etc. In this paper, we study the problem of searching the \underlineB est \underlineC o-located \underlineC ommunity (\BCC) in attributed networks, which returns a community that satisfies the following properties: i) structural cohesiveness: members in the community are densely connected, ii) spatial co-location: members are close to each other, and iii) quality optimality: the community has the best quality in terms of given attributes. The problem can be used in social network user behavior analysis, recommendation systems, disease predication, etc. We first propose an index structure called \DTree to integrate the spatial information, the local structure information, and the attribute information together to accelerate the query processing. Then, based on this index we develop an efficient algorithm. The experimental study conducted on both real and synthetic datasets demonstrate the efficiency and effectiveness of the proposed methods.

In this paper, we develop a neural attentive interpretable recommendation system, named NAIRS. A self-attention network, as a key component of the system, is designed to assign attention weights to interacted items of a user. This attention mechanism can distinguish the importance of the various interacted items in contributing to a user profile. %, and it also provides interpretable recommendations. Based on the user profiles obtained by the self-attention network, NAIRS offers personalized high-quality recommendation. Moreover, it develops visual cues to interpret recommendations. This demo application with the implementation of NAIRS enables users to interact with a recommendation system, and it persistently collects training data to improve the system. The demonstration and experimental results show the effectiveness of NAIRS.

With advancement of various information processing and storage techniques, the scale of digital music collections has been growing at very fast speed during recent decades. To support high-quality content-based retrieval over such a large volume of music data, how to develop indexing structure with good effectiveness, efficiency and scalability becomes an important research issue. However, existing techniques mainly focus on improving query efficiency. Very few approaches have been proposed to address issues related to scalability and accuracy. In this study, we address the problem via introducing a novel indexing technique called effective music indexing framework (EMIF) to facilitate scalable and accurate music retrieval. It is designed based on a “classification-and-indexing” principle and consists of two main functionality modules: (1) music classification—a novel semantic-sensitive classification to identify an input song’s category and (2) indexing module—multiple local indexing structures, one for each semantic category to reduce query response time significantly. In particular, the classification model combining linear discriminative mixture model (LDMM) and advanced score fusion scheme has been applied to estimate category of music accurately. Layered architecture enables EMIF to enjoy superior scalability and efficiency. To evaluate the approach, a set of experimental studies has been carried out using two large music test collections and the results demonstrate various advantages of EMIF over state-of-the-art approaches including efficiency, scalability and effectiveness.

How to minimize the impact of negative users within the maximal set of influenced users? The Influenced Maximization (IM) is important for various applications. However, few studies consider the negative impact of some of the influenced users.We propose a negative-aware influence maximization problem by considering users' negative impact. A novel algorithm is proposed to solve the problem. Experiments on real-world datasets show the proposed algorithm can achieve 70% improvement on average in expected influence compared with rivals.

In this paper, we propose an adversarial process for abstractive text summarization, in which we simultaneously train a generative model G and a discriminative model D. In particular, we build the generator G as an agent of reinforcement learning, which takes the raw text as input and predicts the abstractive summarization. We also build a discriminator which attempts to distinguish the generated summary from the ground truth summary. Extensive experiments demonstrate that our model achieves competitive ROUGE scores with the state-of-the-art methods on CNN/Daily Mail dataset. Qualitatively, we show that our model is able to generate more abstractive, readable and diverse summaries.

This study takes the lead to study the aspect/sentiment-aware abstractive review summarization in domain adaptation scenario. The proposed model CASAS (neural attentive model for Cross-domain Aspect/Sentiment-aware Abstractive review Summarization) leverages domain classification task, working on datasets of both source and target domains, to recognize the domain information of texts and transfer knowledge from source domains to target domains. The extensive experiments on Amazon reviews demonstrate that CASAS outperforms the compared methods in both out-of-domain and in-domain setups.

Review text has been widely studied in traditional tasks such as sentiment analysis and aspect extraction. However, to date, no work is towards the abstractive review summarization that is essential for business organizations and individual consumers to make informed decisions. This work takes the lead to study the aspect/sentiment-aware abstractive review summarization by exploring multi-factor attentions. Specifically, we propose an interactive attention mechanism to interactively learns the representations of context words, sentiment words and aspect words within the reviews, acted as an encoder. The learned sentiment and aspect representations are incorporated into the decoder to generate aspect/sentiment-aware review summaries via an attention fusion network. In addition, the abstractive summarizer is jointly trained with the text categorization task, which helps learn a category-specific text encoder, locating salient aspect information and exploring the variations of style and wording of content with respect to different text categories. The experimental results on a real-life dataset demonstrate that our model achieves impressive results compared to other strong competitors.

Mobile devices generate massive amounts of data that is used to get an insight into the user behavior by enterprise systems. Data privacy is a concern in such systems as users have little control over the data that is generated by them. Blockchain systems offer ways to ensure privacy and security of the user data with the implementation of an access control mechanism. In this demonstration, we present ChainMOB, a mobility analytics application that is built on top of blockchain and addresses the fundamental privacy and security concerns in enterprise systems. Further, the extent of data sharing along with the intended audience is also controlled by the user. Another exciting feature is that user is part of the business model and is incentivized for sharing the personal mobility data. The system also supports queries that can be used in a variety of application domains.

In this paper, we propose a personalized dialogue generation system, which combines reinforcement learning techniques with an attention-based hierarchical recurrent encoderdecoder model. Firstly, we incorporate user-specific information into the decoder to capture user's background information and speaking style. Secondly, we employ reinforcement learning techniques to maximize future reward in dialogue, which enables our system to generate topic-coherent, informative and grammatical responses. Moreover, we propose three types of rewards to characterize good conversations. Finally, we compare the performance of the following reinforcement learning methods in dialogue generation: policy gradient, Q-learning, and actor-critic algorithms. We conduct experiments to verify the effectiveness of the proposed model on two dialogue datasets. Experimental results demonstrate that our model can generate better personalized dialogues for different users. Quantitatively, our method achieves better performance than the state-of-the-art dialogue systems in terms of BLEU score, perplexity, and human evaluation.

Metro systems have become one of the most important public transit services in cities. It is important to understand individual metro passengers' spatio-temporal travel patterns. More specifically, for a specific passenger: what are the temporal patterns? what are the spatial patterns? is there any relationship between the temporal and spatial patterns? are the passenger's travel patterns normal or special? Answering all these questions can help to improve metro services, such as evacuation policy making and marketing. Given a set of massive smart card data over a long period, how to effectively and systematically identify and understand the travel patterns of individual passengers in terms of space and time is a very challenging task. This paper proposes an effective data-mining procedure to better understand the travel patterns of individual metro passengers in Shenzhen, a modern and big city in China. First, we investigate the travel patterns in individual level and devise the method to retrieve them based on raw smart card transaction data, then use statistical-based and unsupervised clustering-based methods, to understand the hidden regularities and anomalies of the travel patterns. From a statistical-based point of view, we look into the passenger travel distribution patterns and find out the abnormal passengers based on the empirical knowledge. From unsupervised clustering point of view, we classify passengers in terms of the similarity of their travel patterns. To interpret the group behaviors, we also employ the bus transaction data. Moreover, the abnormal passengers are detected based on the clustering results. At last, we provide case studies and findings to demonstrate the effectiveness of the proposed scheme.

This paper proposes a novel super-resolution method that exploits the sparse representation and non-local similarity of patches for the effective reconstruction of images. Highresolution images are reconstructed from low resolution observations with an efficient technique based on the alternating direction method of multipliers (ADMM). A robust iterative back-projection approach is used in a post-processing step to remove residual noise and artifacts in the reconstructed image. Experiments on benchmark medical images illustrate the advantage of our method, in terms of PSNR and SSIM, compared to state of the art approaches.

The sparse representation models for image super-resolution have shown great potential in various imaging and vision tasks. However, most of them are challenged by the accuracy issue especially when images are significantly down-sampled. In this paper, we aim to improve the performance of sparse representation. We propose to incorporate a low rank approach into image non-local sparse representation model. To the best of our knowledge, this is the first work to integrate low rank approaches into non-local spare representation for image interpolation. The proposed method can obtain good estimation of sparse coefficients of original images. Experimental results show the effectiveness of our proposed method compared with the state-of-the-art.

Real-world graphs are not always publicly available or sometimes do not meet specific research requirements. These challenges call for generating synthetic networks that follow properties of the real-world networks. Barabási-Albert (BA) is a well-known model for generating scale-free graphs, i.e graphs with power-law degree distribution. In BA model, the network is generated through an iterative stochastic process called preferential attachment. Although BA is highly demanded, due to the inherent complexity of the preferential attachment, this model cannot be scaled to generate billion-node graphs. In this paper, we propose ROLL-tree, a fast in-memory roulette wheel data structure that accelerates the BA network generation process by exploiting the statistical behaviors of the underlying growth model. Our proposed method has the following properties: (a) Fast: It performs +1000 times faster than the state-of-the-art on a single node PC; (b) Exact: It strictly follows the BA model, using an efficient data structure instead of approximation techniques; (c) Generalizable: It can be adapted for other "rich-get-richer" stochastic growth models. Our extensive experiments prove that ROLL-tree can effectively accelerate graph-generation through the preferential attachment process. On a commodity single processor machine, for example, ROLL-tree generates a scale-free graph of 1.1 billion nodes and 6.6 billion edges (the size of Yahoo's Webgraph) in 62 minutes while the state-of-the-art (SA) takes about four years on the same machine.

Information diffusion in social networks is often characterized by huge participating communities and viral cascades of high dynamicity. To observe, summarize, and understand the evolution of dynamic diffusion processes in an informative and insightful way is a challenge of high practical value. However, few existing studies aim to summarize networks for interesting dynamic patterns. Dynamic networks raise new challenges not found in static settings, including time sensitivity, online interestingness evaluation, and summary traceability, which render existing techniques inadequate. We propose dynamic network summarization to summarize dynamic networks with millions of nodes by only capturing the few most interesting nodes or edges overtime. Based on the concepts of diffusion radius and scope, we define interestingness measures for dynamic networks, and we propose OSNet, an online summarization framework for dynamic networks. Efficient algorithms are included in OSNet. We report on extensive experiments with both synthetic and real-life data. The study offers insight into the effectiveness, efficiency, and design properties of OSNet.

Data collection is required to be safe and efficient considering both data privacy and system performance. In this paper, we study a new problem: distributed data sharing with privacy-preserving requirements. Given a data demander requesting data from multiple distributed data providers, the objective is to enable the data demander to access the distributed data without knowing the privacy of any individual provider. The problem is challenged by two questions: how to transmit the data safely and accurately; and how to efficiently handle data streams? As the first study, we propose a practical method, Shadow Coding, to preserve the privacy in data transmission and ensure the recovery in data collection, which achieves privacy preserving computation in a data-recoverable, efficient, and scalable way. We also provide practical techniques to make Shadow Coding efficient and safe in data streams. Extensive experimental study on a large-scale real-life dataset offers insight into the performance of our schema. The proposed schema is also implemented as a pilot system in a city to collect distributed mobile phone data.

The availability of trajectories tracking the geographical locations of people as a function of time offers an opportunity to study human behaviors. In this article, we study rationality from the perspective of user decision on visiting a point of interest (POI) which is represented as a trajectory. However, the analysis of rationality is challenged by a number of issues, for example, how to model a trajectory in terms of complex user decision processes? and how to detect hidden factors that have significant impact on the rational decision making? In this study, we propose Rationality Analysis Model (RAM) to analyze rationality from trajectories in terms of a set of impact factors. In order to automatically identify hidden factors, we propose a method, Collective Hidden Factor Retrieval (CHFR), which can also be generalized to parse multiple trajectories at the same time or parse individual trajectories of different time periods. Extensive experimental study is conducted on three large-scale real-life datasets (i.e., taxi trajectories, user shopping trajectories, and visiting trajectories in a theme park). The results show that the proposed methods are efficient, effective, and scalable. We also deploy a system in a large theme park to conduct a field study. Interesting findings and user feedback of the field study are provided to support other applications in user behavior mining and analysis, such as business intelligence and user management for marketing purposes.

In step with the web being used widely by mobile users, user location is becoming an essential signal in services, including local intent search. Given a large set of spatial web objects consisting of a geographical location and a textual description (e.g., Online business directory entries of restaurants, bars, and shops), how can we find sets of objects that are both spatially and textually relevant to a query? Most of existing studies solve the problem by requiring that all query keywords are covered by the returned objects and then rank the sets by spatial proximity. The needs for identifying sets with more textually relevant objects render these studies inapplicable. We propose locality Search, a query that returns top-k sets of spatial web objects and integrates spatial distance and textual relevance in one ranking function. We show that computing the query is NP-hard, and we present two efficient exact algorithms and one generic approximate algorithm based on greedy strategies for computing the query. We report on findings from an empirical study with three real-life datasets. The study offers insight into the efficiency and effectiveness of the proposed algorithms.

Increasing volumes of geo-referenced data are becoming available. This data includes so-called points of interest that describe businesses, tourist attractions, etc. by means of a geo-location and properties such as a textual description or ratings. We propose and study the efficient implementation of a new kind of query on points of interest that takes into account both the locations and properties of the points of interest. The query takes a result cardinality, a spatial range, and property-related preferences as parameters, and it returns a compact set of points of interest with the given cardinality and in the given range that satisfies the preferences. Specifically, the points of interest in the result set cover so-called allying preferences and are located far from points of interest that possess so-called alienating preferences. A unified result rating function integrates the two kinds of preferences with spatial distance to achieve this functionality. We provide efficient exact algorithms for this kind of query. To enable queries on large datasets, we also provide an approximate algorithm that utilizes a nearest-neighbor property to achieve scalable performance. We develop and apply lower and upper bounds that enable search-space pruning and thus improve performance. Finally, we provide a generalization of the above query and also extend the algorithms to support the generalization. We report on an experimental evaluation of the proposed algorithms using real point of interest data from Google Places for Business that offers insight into the performance of the proposed solutions.

Despite the wealth of research on frequent graph pattern mining, how to efficiently mine the complete set of those with constraints still poses a huge challenge to the existing algorithms mainly due to the inherent bottleneck in the mining paradigm. In essence, mining requests with explicitly-specified constraints cannot be handled in a way that is direct and precise. In this paper, we propose a direct mining framework to solve the problem and illustrate our ideas in the context of a particular type of constrained frequent patterns --- the "skinny" patterns, which are graph patterns with a long backbone from which short twigs branch out. These patterns, which we formally define as l-long δ-skinny patterns, are able to reveal insightful spatial and temporal trajectory patterns in mobile data mining, information diffusion, adoption propagation, and many others.