Search

Online question matching is the process of comparing user queries with system questions to find appropriate answers. This task has become increasingly important with the popularity of knowledge sharing social networks in product search and intelligent Q &A in customer service. Many previous studies have focused on designing complex semantic structures through the questions themselves. In fact, the online user’s queries accumulate a large number of similar sentences, which have been grouped by semantics in the retrieval system. However, how to use these sentences to enhance the understanding of system questions is rarely studied. In this paper, we propose a novel Relation-aware Semantic Enhancement Network (RSEN) model. Specifically, we leverage the labels of the history records to identify different semantically related sentences. Then, we construct an expanded relation network to integrate the representation of different semantic relations. Furthermore, we interact we integrate the features of the system question with the semantically related sentences to augment the semantic information. Finally, we evaluate our proposed RSEN on two publicly available datasets. The results demonstrate the effectiveness of our proposed RSEN method compared to the advanced baselines. [ABSTRACT FROM AUTHOR]

Stored-grain temperature is the most important factor in grain storage. According to the measured data, the temperature in the grain pile can be effectively predicted, which can find problems in advance, reduce grain loss and increase grain quality. Long Short-Term memory (LSTM) can perform better in longer sequences than ordinary RNN. This paper is applied to the analysis of big data of grain storage and the early warning of grain storage temperature. In this paper, the selected LSTM is optimized and the early warning model of grain situation is established, and the analysis steps of the early warning model are given. In order to verify the availability of the improved LSTM network structure, RNN and three variants were used to predict the grain temperature under the same conditions, the prediction effect of the improved CLSTM is better. [ABSTRACT FROM AUTHOR]

In the area of community question answering (CQA), answer selection and answer ranking are two tasks which are applied to help users quickly access valuable answers. Existing solutions mainly exploit the syntactic or semantic correlation between a question and its related answers (Q&A), where the multifacet domain effects in CQA are still underexplored. In this paper, we propose a unified model, enhanced attentive recurrent neural network (EARNN), for both answer selection and answer ranking tasks by taking full advantages of both Q&A semantics and multifacet domain effects (i.e., topic effects and timeliness). Specifically, we develop a serialized long short-term memory to learn the unified representations of Q&A, where two attention mechanisms at either sentence level or word level are designed for capturing the deep effects of topics. Meanwhile, the emphasis of Q&A can be automatically distinguished. Furthermore, we design a time-sensitive ranking function to model the timeliness in CQA. To effectively train EARNN, a question-dependent pairwise learning strategy is also developed. Finally, we conduct extensive experiments on a real-world dataset from Quora. Experimental results validate the effectiveness and interpretability of our proposed EARNN model. [ABSTRACT FROM AUTHOR]

Information source detection is to identify nodes initiating the diffusion process in a network, which has a wide range of applications including epidemic outbreak prevention, Internet virus source identification, and rumor source tracing in social networks. Although it has attracted ever-increasing attention from research community in recent years, existing solutions still suffer from high time complexity and inadequate effectiveness, due to high dynamics of information diffusion and observing just a snapshot of the whole process. To this end, we present a comprehensive study for single information source detection in weighted graphs. Specifically, we first propose a maximum a posteriori (MAP) estimator to detect the information source with other methods as the prior, which ensures our method can be integrated with others naturally. Different from many related works, we exploit both infected nodes and their uninfected neighbors to calculate the effective propagation probability, and then derive the exact formation of likelihood for general weighted graphs. To further improve the efficiency, we design two approximate MAP estimators, namely brute force search approximation (BFSA) and greedy search bound approximation (GSBA), from the perspective of likelihood approximation. BFSA tries to traverse the permitted permutations to directly compute the likelihood, but GSBA exploits a strategy of greedy search to find a surrogate upper bound of the likelihood, and thus avoids the enumeration of permitted permutations. Therefore, detecting with partial nodes and likelihood approximation reduces the computational complexity drastically for large graphs. Extensive experiments on several data sets also clearly demonstrate the effectiveness of our methods on detecting the single information source with different settings in weighted graphs. [ABSTRACT FROM AUTHOR]

Ten years ago, linear models were applied in various domains. Before application of the algorithms, several current studies extracted features presumed to represent parochial markings from the data using engineering techniques. Recently the deep learning domain offered opportunities to directly feed data into the model without any extensive hand-crafted feature engineering techniques. In this paper, the proposed framework does the feature extraction in a non-supervised (i.e. self-supervised) manner using both Contextual Long Short-Term Memory (CLSTM) and Contextual Convolutional Neural Networks (CCNN). We can then concatenate data obtained from the CLSTM and CCNN blocks, feed it into the Attention block, pass it through the Multilayer Perceptron (MLP) block, ultimately passing it through a terminal layer for classification. The task involved here was non-trivial as there is a major challenge in implementing our model to solve the time series classification (TSC) problem: overfitting. We deal with this challenge as follows; firstly, we adjusted the number of neurons in each of the stages. Secondly we introduced dropouts after every layer in each stage of this model. Finally experiments regarding the University of California Riverside (UCR) dataset indicates the model's superiority. [ABSTRACT FROM AUTHOR]

Traditionally, research about social user profiling assumes that users share some similar interests with their followees. However, it lacks the studies on what topic and to what extent their interests are similar. Our study in online sharing sites reveals that besides shared interests between followers and followees, users do maintain some individual interests which differ from their followees. Thus, for better social user profiling we need to discern individual interests (capturing the uniqueness of users) and shared interests (capturing the commonality of neighboring users) of the users in the connected world. To achieve this, we extend the matrix factorization model by incorporating both individual and shared interests, and also learn the multi-faceted similarities unsupervisedly. The proposed method can be applied to many applications, such as rating prediction, item level social influence maximization and so on. Experimental results on real-world datasets show that our work can be applied to improve the performance of social rating. Also, it can reveal some interesting findings, such as who likes the 'controversial' items most, and who is the most influential in attracting their followers to rate an item. [ABSTRACT FROM AUTHOR]

Recent advances in mobile devices and their sensing capabilities have enabled the collection of rich contextual information and mobile device usage records through the device logs. These context-rich logs open a venue for mining the personal preferences of mobile users under varying contexts and thus enabling the development of personalized context-aware recommendation and other related services, such as mobile online advertising. In this article, we illustrate how to extract personal context-aware preferences from the context-rich device logs, or context logs for short, and exploit these identified preferences for building personalized context-aware recommender systems. A critical challenge along this line is that the context log of each individual user may not contain sufficient data for mining his or her context-aware preferences. Therefore, we propose to first learn common context-aware preferences from the context logs of many users. Then, the preference of each user can be represented as a distribution of these common context-aware preferences. Specifically, we develop two approaches for mining common context-aware preferences based on two different assumptions, namely, context-independent and context-dependent assumptions, which can fit into different application scenarios. Finally, extensive experiments on a real-world dataset show that both approaches are effective and outperform baselines with respect to mining personal context-aware preferences for mobile users. [ABSTRACT FROM AUTHOR]

Recent advances in mobile devices and their sensing capabilities have enabled the collection of rich contextual information and mobile device usage records through the device logs. These context-rich logs open a venue for mining the personal preferences of mobile users under varying contexts and thus enabling the development of personalized context-aware recommendation and other related services, such as mobile online advertising. In this article, we illustrate how to extract personal context-aware preferences from the context-rich device logs, or context logs for short, and exploit these identified preferences for building personalized context-aware recommender systems. A critical challenge along this line is that the context log of each individual user may not contain sufficient data for mining his or her context-aware preferences. Therefore, we propose to first learn common context-aware preferences from the context logs of many users. Then, the preference of each user can be represented as a distribution of these common context-aware preferences. Specifically, we develop two approaches for mining common context-aware preferences based on two different assumptions, namely, context-independent and context-dependent assumptions, which can fit into different application scenarios. Finally, extensive experiments on a real-world dataset show that both approaches are effective and outperform baselines with respect to mining personal context-aware preferences for mobile users. [ABSTRACT FROM AUTHOR]

The problem of expert finding targets on identifying experts with special skills or knowledge for some particular knowledge categories, i.e. knowledge domains, by ranking user authority. In recent years, this problem has become increasingly important with the popularity of knowledge sharing social networks. While many previous studies have examined authority ranking for expert finding, they have a focus on leveraging only the information in the target category for expert finding. It is not clear how to exploit the information in the relevant categories of a target category for improving the quality of authority ranking. To that end, in this paper, we propose an expert finding framework based on the authority information in the target category as well as the relevant categories. Along this line, we develop a scalable method for measuring the relevancies between categories through topic models, which takes consideration of both content and user interaction based category similarities. Also, we provide a topical link analysis approach, which is multiple-category-sensitive, for ranking user authority by considering the information in both the target category and the relevant categories. Finally, in terms of validation, we evaluate the proposed expert finding framework in two large-scale real-world data sets collected from two major commercial Question Answering (Q&A) web sites. The results show that the proposed method outperforms the baseline methods with a significant margin. [ABSTRACT FROM AUTHOR]

Recent years have witnessed an increased interest in recommender systems. Despite significant progress in this field, there still remain numerous avenues to explore. Indeed, this paper provides a study of exploiting online travel information for personalized travel package recommendation. A critical challenge along this line is to address the unique characteristics of travel data, which distinguish travel packages from traditional items for recommendation. To that end, in this paper, we first analyze the characteristics of the existing travel packages and develop a tourist-area-season topic (TAST) model. This TAST model can represent travel packages and tourists by different topic distributions, where the topic extraction is conditioned on both the tourists and the intrinsic features (i.e., locations, travel seasons) of the landscapes. Then, based on this topic model representation, we propose a cocktail approach to generate the lists for personalized travel package recommendation. Furthermore, we extend the TAST model to the tourist-relation-area-season topic (TRAST) model for capturing the latent relationships among the tourists in each travel group. Finally, we evaluate the TAST model, the TRAST model, and the cocktail recommendation approach on the real-world travel package data. Experimental results show that the TAST model can effectively capture the unique characteristics of the travel data and the cocktail approach is, thus, much more effective than traditional recommendation techniques for travel package recommendation. Also, by considering tourist relationships, the TRAST model can be used as an effective assessment for travel group formation. [ABSTRACT FROM AUTHOR]

Abstract: Recent years have witnessed a considerable surge of interest in the multi-label learning problem. It has been shown that a key factor for a successful multi-label learning algorithm is to effectively exploit relations between labels. However, most of the previous work exploiting label relations focuses on pairwise relations. To handle the situations where there are intrinsic correlations among multiple labels, in this paper, we propose a generative model, Labeled Four-Level Pachinko Allocation Model (L-F-L-PAM), to capture correlations among multiple labels. In our approach of multi-label learning on text data, we apply the proposed model for inferring the training data and the standard Four-Level Pachinko Allocation Model for the test data. Furthermore, we propose a pruned Gibbs Sampling algorithm in the test stage to reduce the inference time. Finally, extensive experiments have been performed to validate the effectiveness and efficiency of our new approach. The results demonstrate significant improvements of our model over Labeled LDA (L-LDA) and superiority in terms of both effectiveness and computational efficiency over other high-performing multi-label learning methods. [Copyright &y& Elsevier]

Recommender systems suggest a few items from many possible choices to the users by understanding their past behaviors. In these systems, the user behaviors are influenced by the hidden interests of the users. Learning to leverage the information about user interests is often critical for making better recommendations. However, existing collaborative-filtering-based recommender systems are usually focused on exploiting the information about the user's interaction with the systems; the information about latent user interests is largely underexplored. To that end, inspired by the topic models, in this paper, we propose a novel collaborative-filtering-based recommender system by user interest expansion via personalized ranking, named iExpand. The goal is to build an item-oriented model-based collaborative-filtering framework. The iExpand method introduces a three-layer, user–interests–item, representation scheme, which leads to more accurate ranking recommendation results with less computation cost and helps the understanding of the interactions among users, items, and user interests. Moreover, iExpand strategically deals with many issues that exist in traditional collaborative-filtering approaches, such as the overspecialization problem and the cold-start problem. Finally, we evaluate iExpand on three benchmark data sets, and experimental results show that iExpand can lead to better ranking performance than state-of-the-art methods with a significant margin. [ABSTRACT FROM AUTHOR]

Abstract: Traditional term weighting schemes in text categorization, such as TF-IDF, only exploit the statistical information of terms in documents. Instead, in this paper, we propose a novel term weighting scheme by exploiting the semantics of categories and indexing terms. Specifically, the semantics of categories are represented by senses of terms appearing in the category labels as well as the interpretation of them by WordNet. Also, the weight of a term is correlated to its semantic similarity with a category. Experimental results on three commonly used data sets show that the proposed approach outperforms TF-IDF in the cases that the amount of training data is small or the content of documents is focused on well-defined categories. In addition, the proposed approach compares favorably with two previous studies. [Copyright &y& Elsevier]

Alternative splicing of the primary RNA transcript is a common mechanism for generating protein diversity. A model system was developed to study this process in vitro that is useful for evaluation of splicing of transcripts expressed in cells that do not grow well in culture. The system was used to analyze skipping of exon 4 of the amelogenin message, normally expressed in ameloblast cells for a short interval during tooth enamel development. Amelogenins are highly conserved proteins resulting from extensive alternative splicing, with domains involved in a range of functions, including mineral formation and intercellular signaling. In the bovine gene, the very short intron 4 was predicted to inhibit inclusion of exon 4, because in murine ameloblasts, exon 4 is detectably included in mRNA, and intron 4 is longer than the bovine counterpart. Bovine intron 4 was lengthened, and this size increase enhanced exon 4 inclusion sixfold to eightfold, although splice site selection was inaccurate. Intron length, therefore, is not the sole determinant controlling amelogenin exon 4 inclusion, and cis-acting inhibitory elements may also be involved in exon skipping. This vector system allows evaluation of splicing of a tissue-specific RNA by focusing on exons of interest through transfection of heterologous cultured cells without complications attributable to background transcription of the gene being evaluated. [ABSTRACT FROM AUTHOR]

The advent of large language models marks a revolutionary breakthrough in artificial intelligence. With the unprecedented scale of training and model parameters, the capability of large language models has been dramatically improved, leading to human-like performances in understanding, language synthesizing, common-sense reasoning, etc. Such a major leap forward in general AI capacity will fundamentally change the pattern of how personalization is conducted. For one thing, it will reform the way of interaction between humans and personalization systems. Instead of being a passive medium of information filtering, like conventional recommender systems and search engines, large language models present the foundation for active user engagement. On top of such a new foundation, users’ requests can be proactively explored, and users’ required information can be delivered in a natural, interactable, and explainable way. For another thing, it will also considerably expand the scope of personalization, making it grow from the sole function of collecting personalized information to the compound function of providing personalized services. By leveraging large language models as a general-purpose interface, the personalization systems may compile user’s requests into plans, calls the functions of external tools (e.g., search engines, calculators, service APIs, etc.) to execute the plans, and integrate the tools’ outputs to complete the end-to-end personalization tasks. Today, large language models are still being rapidly developed, whereas the application in personalization is largely unexplored. Therefore, we consider it to be right the time to review the challenges in personalization and the opportunities to address them with large language models. In particular, we dedicate this perspective paper to the discussion of the following aspects: the development and challenges for the existing personalization system, the newly emerged capabilities of large language models, and the potential ways of making use of large language models for personalization. [ABSTRACT FROM AUTHOR]

The article introduces UniTRep, a unified representation learning framework for career mobility analysis, aiming to address the lack of cross-scale interactions between talents and the labor market. Topics include leveraging trajectory hypergraphs for organizing career mobility patterns, attentively propagating market contextual features into individual trajectory modeling, and optimizing learning with self-supervised strategies.

In the real estate industry, Days on Market (DOM) is one of the most important attribute that is normally used to appraise real estate properties. In the academic sector, DOM is seemingly attracting a lot of researchers. DOM can be define as the length of time (i.e. in days) a real estate listing takes in passive market. In our paper, a novel hybrid neural network model is proposed to solve DOM prediction problem. Our proposed model extracts features using both CNN-based Attention (CNNA), and Bidirectional LSTM (BLSTM) modules. Furthermore, we concatenate their outputs and pass the results through a prediction (MLP) block, for predictions to be made. In implementing our model, overfitting was experienced as a challenge. In order to combat overfitting in our network we introduce Dropout layers in almost all the modules. Moreover, we present confidence intervals for four attributes in our dataset by using either percentile bootstrap confidence interval (CI) or percentile bias corrected accelerated (B C a ) bootstrap CI, depending on the estimated distribution of an attribute. Finally, we appraise our model by experimenting with dataset of a famous real estate agency in Shanghai. The experimental outcomes clearly prove the superiority of the projected approach for solving DOM prediction problem. [ABSTRACT FROM AUTHOR]

Because of strict response-time constraints, efficiency of top-k recommendation is crucial for real-world recommender systems. Locality sensitive hashing and index-based methods usually store both index data and item feature vectors in main memory, so they handle a limited number of items. Hashing-based recommendation methods enjoy low memory cost and fast retrieval of items, but suffer from large accuracy degradation. In this paper, we propose product Quantized Collaborative Filtering (pQCF) for better trade-off between efficiency and accuracy. pQCF decomposes a joint latent space of users and items into a Cartesian product of low-dimensional subspaces, and learns clustered representation within each subspace. A latent factor is then represented by a short code, which is composed of subspace cluster indexes. A user’s preference for an item can be efficiently calculated via table lookup. We then develop block coordinate descent for efficient optimization and reveal the learning of latent factors is seamlessly integrated with quantization. We further investigate an asymmetric pQCF, dubbed as QCF, where user latent factors are not quantized and shared across different subspaces. The extensive experiments with 6 real-world datasets show that pQCF significantly outperforms the state-of-the-art hashing-based CF and QCF increases recommendation accuracy compared to pQCF. [ABSTRACT FROM AUTHOR]

Binary representation of users and items can dramatically improve efficiency of recommendation and reduce size of recommendation models. However, learning optimal binary codes for them is challenging due to binary constraints, even if squared loss is optimized. In this article, we propose a general framework for discrete matrix factorization based on discrete optimization, which can 1) optimize multiple loss functions; 2) handle both explicit and implicit feedback datasets; and 3) take auxiliary information into account without any hyperparameters. To tackle the challenging discrete optimization problem, we propose block coordinate descent based on semidefinite relaxation of binary quadratic programming. We theoretically show that it is equivalent to discrete coordinate descent when only one coordinate is in each block. We extensively evaluate the proposed algorithms on eight real-world datasets. The results of evaluation show that they outperform the state-of-the-art baselines significantly and that auxiliary information of items improves recommendation performance. For better showing the advantages of binary representation, we further propose a two-stage recommender system, consisting of an item-recalling stage and a subsequent fine-ranking stage. Its extensive evaluation shows hashing can dramatically accelerate item recommendation with little degradation of accuracy. [ABSTRACT FROM AUTHOR]

Recent studies have shown that amphoteric regulatory protein (AREG), a member of the epidermal growth factor (EGF) family, is expressed in many cancers and is an independent prognostic indicator for patients with pancreatic cancer, but whether AREG is regulated at the epigenetic level to promote the development of pancreatic cancer (PC) has not been elucidated. Our results support the notion that AREG is overexpressed in pancreatic cancer tissues and cell lines. Functionally, the deletion of AREG impedes pancreatic cancer (PC) cell proliferation, migration, and invasion. In addition, we identified and validated that methyltransferase-like 3 (METTL3) induced the m6A modification on AREG and facilitated the stability of AREG mRNA after sequencing. Additionally, we obtained experimental evidence that miR-33a-3p targets and inhibits METTL3 from taking action, as predicted by using the miRDB and RNAinter. Remediation experiments showed that miR-33a-3p inhibits PC progression through METTL3. In summary, this research reveals that miR-33a-3p inhibits m6A-induced stabilization of AREG by targeting METTL3, which plays a key role in the aggressive progression of PC. AREG could be a potential target for PC treatment. [ABSTRACT FROM AUTHOR]

Recommender systems are often susceptible to well-crafted fake profiles, leading to biased recommendations. Among existing defense methods, data-processing-based methods inevitably exclude normal samples, while model-based methods struggle to enjoy both generalization and robustness. To this end, we suggest integrating data processing and the robust model to propose a general framework, Triple Cooperative Defense (TCD), which employs three cooperative models that mutually enhance data and thereby improve recommendation robustness. Furthermore, Considering that existing attacks struggle to balance bi-level optimization and efficiency, we revisit poisoning attacks in recommender systems and introduce an efficient attack strategy, Co-training Attack (Co-Attack), which cooperatively optimizes the attack optimization and model training, considering the bi-level setting while maintaining attack efficiency. Moreover, we reveal a potential reason for the insufficient threat of existing attacks is their default assumption of optimizing attacks in undefended scenarios. This overly optimistic setting limits the potential of attacks. Consequently, we put forth a Game-based Co-training Attack (GCoAttack), which frames the proposed CoAttack and TCD as a game-theoretic process, thoroughly exploring CoAttack's attack potential in the cooperative training of attack and defense. Extensive experiments on three real datasets demonstrate TCD's superiority in enhancing model robustness. Additionally, we verify that the two proposed attack strategies significantly outperform existing attacks, with game-based GCoAttack posing a greater poisoning threat than CoAttack. [ABSTRACT FROM AUTHOR]

The efficiency of communication across workers is a significant factor that affects the performance of federated learning. Though periodic communication strategy is applied to reduce communication rounds in training, the communication cost is still high when the training data distributions are not independently and identically distributed (non-IID) which is common in federated learning. Recently, some works introduce variance reduction to eliminate the effect caused by non-IID data among workers. Nevertheless the provable optimal communication complexity O (log (S T)) and convergence rate O (1 / (S T)) cannot be achieved simultaneously, where S denotes the number of sampled workers in each round and T is the number of iterations. To deal with this dilemma, we propose an optimization algorithm SQUARFA that adopts stagewise training framework coupling with variance reduction and uses a quick-start phase in each loop. Theoretical results show that SQUARFA achieves both optimal convergence rate and communication complexity for both strongly convex objectives and non-convex objectives under PL condition, thus fills the gap mentioned above. Then, a variant of SQUARFA yields the optimal theoretical results for general non-convex objectives. We further extend the technique in SQUARFA to the large batch setting and achieve optimal communication complexity. Experimental results demonstrate the superiority of the proposed algorithms. [ABSTRACT FROM AUTHOR]

Recent studies have shown that amphoteric regulatory protein (AREG), a member of the epidermal growth factor (EGF) family, is expressed in many cancers and is an independent prognostic indicator for patients with pancreatic cancer, but whether AREG is regulated at the epigenetic level to promote the development of pancreatic cancer (PC) has not been elucidated. Our results support the notion that AREG is overexpressed in pancreatic cancer tissues and cell lines. Functionally, the deletion of AREG impedes pancreatic cancer (PC) cell proliferation, migration, and invasion. In addition, we identified and validated that methyltransferase-like 3 (METTL3) induced the m6A modification on AREG and facilitated the stability of AREG mRNA after sequencing. Additionally, we obtained experimental evidence that miR-33a-3p targets and inhibits METTL3 from taking action, as predicted by using the miRDB and RNAinter. Remediation experiments showed that miR-33a-3p inhibits PC progression through METTL3. In summary, this research reveals that miR-33a-3p inhibits m6A-induced stabilization of AREG by targeting METTL3, which plays a key role in the aggressive progression of PC. AREG could be a potential target for PC treatment. [ABSTRACT FROM AUTHOR]

Recent decades have witnessed the rapid growth of educational data mining (EDM), which aims at automatically extracting valuable information from large repositories of data generated by or related to people’s learning activities in educational settings. One of the key EDM tasks is cognitive modelling with examination data, and cognitive modelling tries to profile examinees by discovering their latent knowledge state and cognitive level (e.g. the proficiency of specific skills). However, to the best of our knowledge, the problem of extracting information from both objective and subjective examination problems to achieve more precise and interpretable cognitive analysis remains underexplored. To this end, we propose a fuzzy cognitive diagnosis framework (FuzzyCDF) for examinees’ cognitive modelling with both objective and subjective problems. Specifically, to handle the partially correct responses on subjective problems, we first fuzzify the skill proficiency of examinees. Then we combine fuzzy set theory and educational hypotheses to model the examinees’ mastery on the problems based on their skill proficiency. Finally, we simulate the generation of examination score on each problem by considering slip and guess factors. In this way, the whole diagnosis framework is built. For further comprehensive verification, we apply our FuzzyCDF to three classical cognitive assessment tasks, i.e., predicting examinee performance, slip and guess detection, and cognitive diagnosis visualization. Extensive experiments on three real-world datasets for these assessment tasks prove that FuzzyCDF can reveal the knowledge states and cognitive level of the examinees effectively and interpretatively. [ABSTRACT FROM AUTHOR]

Online education brings more possibilities for personalized learning, in which identifying the cognitive state of learners is conducive to better providing learning services. Cognitive diagnosis is an effective measurement to assess the cognitive state of students through response data of answering the problems(e.g., right or wrong). Generally, the cognitive diagnosis framework includes the mastery of skills required by a specified problem and the aggregation of skills. The current multi-skill aggregation methods are mainly divided into conjunctive and compensatory methods and generally considered that each skill has the same effect on the correct response. However, in practical learning situations, there may be more complex interactions between skills, in which each skill has different weight impacting the final result. To this end, this paper proposes a generalized multi-skill aggregation method based on the Sugeno integral (SI-GAM) and introduces fuzzy measures to characterize the complex interactions between skills. We also provide a new idea for modeling multi-strategy problems. The cognitive diagnosis process is implemented by a more general and interpretable aggregation method. Finally, the feasibility and effectiveness of the model are verified on synthetic and real-world datasets. [ABSTRACT FROM AUTHOR]

One-class collaborative filtering (OCCF) problems are ubiquitous in real-world recommendation systems, such as news recommendation, but suffer from data sparsity and lack of negative items. To address the challenge, the state-of-the-art algorithm assigns uninteracted items with smaller weights of being negative and performs low-rank approximation over the user-item interaction matrix. However, the prior ratings are usually suggested to be zero but may not be well-defined. To avert the direct utilization of prior ratings for uninteracted items, we propose a novel ranking-based implicit regularizer by hypothesizing that users’ preference scores for uninteracted items should not deviate a lot from each other. The regularizer is then used in a ranking-based OCCF framework to penalize large differences of preference scores between uninteracted items. To efficiently optimize model parameters in this framework, we develop the scalable alternating least square algorithm and coordinate descent algorithm, whose time complexity is linearly proportional to the data size. Finally, we extensively evaluate the proposed algorithms on six public real-world datasets. The results show that the proposed regularizer significantly improves the recommendation quality of ranking-based OCCF algorithms, such as BPRMF and RankALS. Moreover, the ranking-based framework with the proposed regularizer outperforms the state-of-the-art recommendation algorithms for implicit feedback. [ABSTRACT FROM AUTHOR]

As one of the important psychological stress reactions, Micro-expressions (MEs) are spontaneous and subtle facial movements, which usually occur in a high-stake situation and can reveal genuine human feelings and cognition. ME, Recognition (MER) has essential applications in many fields such as lie detection, criminal investigation, and psychological healing. However, due to the challenges of learning discriminative ME features via fleeting facial subtle reactions as well as the shortage of available MEs data, this research topic is still far from well-studied. To this end, in this paper, we propose a deep prototypical learning framework, namely ME-PLAN, with a local attention mechanism for the MER problem. Specifically, ME-PLAN consists of two components, i.e., a 3D residual prototypical network and a local-wise attention module, where the former aims to learn the precise ME feature prototypes through expression-related knowledge transfer and episodic training, and the latter could facilitate the attention to the local facial movements. Furthermore, to alleviate the dilemma that most MER methods need to depend on manually annotated apex frames, we propose an apex frame spotting method with Unimodal Pattern Constrained (UPC) and further extract ME key-frames sequences based on the detected apex frames to train our proposed ME-PLAN in an end-to-end manner. Finally, through extensive experiments and interpretable analysis regarding the apex frame spotting and MER on composite-database, we demonstrate the superiority and effectiveness of the proposed methods. • We explore deep prototypical learning for the MER problem, and try to address three challenges in this topic. • We propose a novel deep prototypical learning framework, namely ME-PLAN, using a 3D residual prototypical network with episodic training and a local-wise attention module to learn precise ME representation. • A novel apex frame spotting method based on Unimodal Pattern Constraint is proposed to effectively eliminate noise interference and accurately locate apex frames. • Extensive experimental results with a comparison of state-of-the-art methods have demonstrated the effectiveness of our ME-PLAN on apex frame spotting and MER tasks. [ABSTRACT FROM AUTHOR]

Collaborative filtering (CF) models offer users personalized recommendations by measuring the relevance between the active user and each individual candidate item. Following this idea, user-based collaborative filtering (UCF) usually selects the local popular items from the like-minded neighbor users. However, these traditional relevance-based models only consider the individuals (i.e., each neighbor user and candidate item) separately during neighbor set selection and recommendation set generation, thus usually incurring highly similar recommendations that lack diversity. While many researchers have recognized the importance of diversified recommendations, the proposed solutions either needed additional semantic information of items or decreased accuracy in this process. In this article, we describe how to generate both accurate and diversified recommendations from a new perspective. Along this line, we first introduce a simple measure of coverage that quantifies the usefulness of the whole set, that is, the neighbor userset and the recommended itemset as a complete entity. Then we propose a recommendation framework named REC that considers both traditional relevance-based scores and the new coverage measure based on UCF. Under REC, we further prove that the goals of maximizing relevance and coverage measures simultaneously in both the neighbor set selection step and the recommendation set generation step are NP-hard. Luckily, we can solve them effectively and efficiently by exploiting the inherent submodular property. Furthermore, we generalize the coverage notion and the REC framework from both a data perspective and an algorithm perspective. Finally, extensive experimental results on three real-world datasets show that the REC-based recommendation models can naturally generate more diversified recommendations without decreasing accuracy compared to some state-of-the-art models. [ABSTRACT FROM AUTHOR]

The implicit feedback based collaborative filtering (CF) has attracted much attention in recent years, mainly because users implicitly express their preferences in many real-world scenarios. The current mainstream pairwise methods optimize the Area Under the Curve (AUC) and are empirically proved to be helpful to exploit binary relevance data, but lead to either not address the ranking problem, or not specifically focus on top- $k$ k recommendation. Although there exists the listwise method maximizes the Mean Reciprocal Rank (MRR), it has low efficiency and is not particularly adequate for general implicit feedback situations. To that end, in this paper, we propose a new framework, namely Collaborative List-and-Pairwise Filtering (CLAPF), which aims to introduce pairwise thinking into listwise methods. Specifically, we smooth another well-known rank-biased measure called Mean Average Precision (MAP), and respectively combine two rank-biased metrics (MAP, MRR) with the pairwise objective function to capture the performance of top- $k$ k recommendation. Furthermore, the sampling scheme for CLAPF is discussed to accelerate the convergence speed. Our CLAPF framework is a new hybrid model that provides an idea of utilizing rank-biased measures in a pairwise way on implicit feedback. Empirical studies demonstrated CLAPF outperforms state-of-the-art approaches on real-world datasets. [ABSTRACT FROM AUTHOR]

With the prevalence of GPS-enabled smart phones, Location Based Social Network (LBSN) has emerged and become a hot research topic during the past few years. As one of the most important components in LBSN, Points-of-Interests (POIs) has been extensively studied by both academia and industry, yielding POI recommendations to enhance user experience in exploring the city. In conventional methods, rating vectors for both users and POIs are utilized for similarity calculation, which might yield inaccuracy due to the differences of user biases. In our opinion, the rating values themselves do not give exact preferences of users, however the numeric order of ratings given by a user within a certain period provides a hint of preference order of POIs by such user. Firstly, we propose an approach to model users preference by employing utility theory. Secondly, We devise a collection-wise learning method over partial orders through an effective stochastic gradient descent algorithm. We test our model on two real world datasets, i.e., Yelp and TripAdvisor, by comparing with some state-of-the-art approaches including PMF and several user preference modeling methods. In terms of MAP and Recall, we averagely achieve 15% improvement with regard to the baseline methods. The results show the significance of comparative choice in a certain time window and show its superiority to the existing methods. [ABSTRACT FROM AUTHOR]

The graph-based multi-view methods have achieved state-of-the-art results in 3D shape analysis tasks by taking advantage of graph convolutional networks (GCN) to process discrete data. However, the homogeneity of the traditional GCN aggregation operator leads to a problem in aggregating neighborhood information, i.e., if several views have the same neighbors, the same node embeddings will be generated, resulting in feature redundancy. To address this problem, we propose a Multi-view Path Aggregation Graph Network (MVPNet) for 3D shape analysis, which aims to extract a particular path from a graph composed of multiple views and aggregate it into an effective 3D shape descriptor. Specifically, we first extract a path in the graph through dynamic walking, and update the path status while searching for new nodes during the walking. Then we embed the position information of the nodes in the order of the nodes in the path. Finally, we propose to aggregate the features of a path employing a Path Transformer that is capable of handling ordered sequences. A path contains richer semantic and structural information than a traditional subgraph. To demonstrate the effectiveness of our proposed method, we conduct extensive experiments on three benchmark datasets, namely ModelNet, ShapeNetCore55 and MCB, and these experiments prove that the method outperforms the current methods in 3D shape classification and retrieval tasks. • We present the first path aggregation graph network for multi-view feature fusion. • A novel dynamic walking strategy is proposed to generate unique paths for 3D shapes. • We propose a ViT-based Path Transformer to aggregate path information. [ABSTRACT FROM AUTHOR]

Traffic flow prediction, a crucial application of intelligent transportation systems (ITS), has become an increasingly prevalent research topic. However, existing models that achieved high prediction accuracy on selected metrics may suffer from time delay in prediction curves which has been rarely explored. These models may produce seemingly accurate but invalid predictions by merely tracking and replicating previous true values. To address this anomaly, we propose a highly interpretable prediction mechanism, the Incremental Output Decomposition Recurrent Neural Network (IODRNN). We also introduce a new metric called Shift Divergence Difference (SDD) to assess the degree of latency of the overall sequence and evaluate the effectiveness of IODRNN in reducing the delay phenomenon. Our experimental results using real-world GNSS traffic data show that IODRNN has the smallest degree of latency and improves MAE and RMSE by 16.8% and 17.4% on average, respectively, over most contrast models. Our study presents an effective approach to evaluate prediction latency, ensuring validity and robustness in traffic prediction. [ABSTRACT FROM AUTHOR]

Abstract: A dynamic voltage scaling technique provides the capability for processors to adjust the speed and control the energy consumption. We study the pessimistic accelerate model where the acceleration rate of the processor speed is at most and jobs cannot be executed during the speed transition period. The objective is to find a min-energy (optimal) schedule that finishes every job within its deadline. The job set we study in this paper is aligned jobs where earlier released jobs have earlier deadlines. We start by investigating a special case where all jobs have a common arrival time and design an algorithm to compute the optimal schedule based on some nice properties of the optimal schedule. Then, we study the general aligned jobs and obtain an algorithm to compute the optimal schedule by using the algorithm for the common arrival time case as a building block. Because our algorithm relies on the computation of the optimal schedule in the ideal model (), in order to achieve complexity, we improve the complexity of computing the optimal schedule in the ideal model for aligned jobs from the currently best known to . [Copyright &y& Elsevier]

Abstract: In this paper, we study the scheduling problem of jobs with multiple active intervals. Each job in the problem instance has disjoint active time intervals where it can be executed and a workload characterized by the required number of CPU cycles. Previously, people studied multiple interval job scheduling problem where each job must be assigned enough CPU cycles in one of its active intervals. We study a different practical version where the partial work done by the end of an interval remains valid and each job is considered finished if total CPU cycles assigned to it in all its active intervals reach the requirement. The goal is to find a feasible schedule that minimizes energy consumption. By adapting the algorithm for single interval jobs proposed in Yao, Demers and Shenker (1995) , one can still obtain an optimal schedule. However, the two phases in that algorithm (critical interval finding and scheduling the critical interval) can no longer be carried out directly. We present polynomial time algorithms to solve the two phases for jobs with multiple active intervals and therefore can still compute the optimal schedule in polynomial time. [Copyright &y& Elsevier]

Abstract: We study the optimal structure for the group broadcast problem where the key tree model is extensively used. The objective is usually to find an optimal key tree to minimize the cost based on certain assumptions. Under the assumption that members arrive in the initial setup period and only member deletions are allowed after that period, previous works show that when only considering the deletion cost, the optimal tree can be computed in time. In this paper, we first prove a semi-balance property for the optimal tree and use it to reduce the running time from to multiplications of -bit integers. Then we study the optimal tree structure when insertion cost is also considered. We show that the optimal tree is such a tree where any internal node has degree at most 7 and children of nodes with degree not equal to 2 or 3 are all leaves. Based on this result we give a dynamic programming algorithm with time to compute the optimal tree. [Copyright &y& Elsevier]

Ontology plays an important role in semantic Web technology since it can effectively represent the domain knowledge. We develop a novel framework for automatically generating the domain knowledge by analyzing different Web sites in a given domain. The idea of our approach is to consider two kinds of information from the Web sites. The first kind of information is the text fragments corresponding to the concepts in the ontology. The other kind of information is the header labels corresponding to the concepts. We design a method for generating the domain ontology by measuring the similarity between the concepts in different Web sites. We have conducted extensive experiments to demonstrate the effectiveness of our approach. [ABSTRACT FROM AUTHOR]

Patent classification has long been treated as a crucial task to support related services. Though large efforts have been made on the automatic patent classification task, those prior arts mainly focus on mining textual information such as titles and abstracts. Meanwhile, few of them pay attention to the meta data, e.g., the inventors and the assignee company, and the potential correlation via the metadata-based graph has been largely ignored. To that end, in this paper, we develop a new paradigm for patent classification task in the perspective of multi-view patent graph analysis and then propose a novel framework called Patent2vec to learn low-dimensional representations of patents for patent classification. Specifically, we first employ the graph representation learning on individual graphs, so that view-specific representations will be learned by capturing the network structure and side information. Then, we propose a view enhancement module to enrich single view representations by exploiting cross-view correlation knowledge. Afterward, we deploy an attention-based multi-view fusion method to get refined representations for each patent and further design a view alignment module to constraint final fused representation in a relational embedding space which can preserve latent relational information. Empirical results demonstrate that our model not only improves the classification accuracy but also improves the interpretability of classifying patents reflected in the multi-source data. [ABSTRACT FROM AUTHOR]

• This paper proposed a novel two-stage spatiotemporal features learning method based on a Siamese 3D CNN for MEs recognition. • The focal loss was introduced to address the problem of inefficient model training caused by the class imbalance in the MEs datasets. • An adaptive construction method based on adaptive CNN was proposed to construct the key-frames sequence to summarize the original MEs. Micro-expressions (MEs) are spontaneous and involuntary facial subtle reactions which often reveal the genuine emotions within human beings. Recognizing MEs automatically is becoming increasingly crucial for many areas such as diagnosis and security. However, the short time duration and low spatial intensity of MEs pose great challenges for accurately recognizing them. Additionally, the lack of sufficient and balanced spontaneous MEs data also makes this problem even harder to solve, and some adaptive modeling strategies have been quite urgent recently. To this end, this paper draws inspirations from few-shot learning to propose a novel two-stage learning (i.e., prior learning and target learning) method based on a siamese 3D convolutional neural network for MEs recognition (MERSiamC3D). Specifically, in the prior learning stage, the proposed MERSiamC3D is used to extract the generic features of MEs. In the target learning stage, the structure and parameters of the MERSiamC3D will be carefully adjusted and the Focal Loss is adopted for high-level features learning. Afterwards, in order to effectively retain the spatiotemporal information of the original MEs video, an adaptive construction method based on adaptive convolutional neural network is proposed to construct the key-frames sequence to summarize the original MEs video, which is able to help drop the redundant frames and relatively highlight the movement of the apex frame. Then, the new key-frames are taken as the input of the two-stage learning method. Finally, through extensive evaluations and experiments on three publically available MEs datasets, the proposed method in this work could outperform traditional methods and other deep learning baselines, which provides a novel insight on how to leverage scarce data for MEs recognition. [ABSTRACT FROM AUTHOR]

Phase-shifting profilometry (PSP) via sequential fringe projection is applicable for high-accuracy 3-D measurements of static objects. However, moving objects cannot be static or quasi-static during one projection cycle, thereby causing surface misalignment and phase errors in 3-D measurements. This article proposes a motion-estimated PSP based on the fringe-oriented synthetic phase correlation algorithm (FO-SPCA), which achieves automatic 3-D measurements of objects in nonuniform 2-D motion. To robustly estimate the motion of objects among fringe images, the proposed fringe-oriented feature extraction technique obtains the surface features by eliminating the modulation influence of sinusoidal illumination. Furthermore, the FO-SPCA is developed to accurately detect the subpixel displacements of moving objects according to object surface features. With the subpixel displacements, the alignment procedure is completed through a reverse 2-D transformation and then the phase errors are compensated accordingly. The simulation and experimental results verify the feasibility and robustness of the motion-estimated PSP for the in situ 3-D measurements of nonuniform moving objects. [ABSTRACT FROM AUTHOR]

Codewords used to determine fringe orders are significant for absolute phase retrieval based on the phase-coding method. However, because of inevitable interference from image blurring, image noise, and system nonlinearity, it is challenging to significantly increase the number of codewords for high-accuracy phase retrieval. To address this challenge, this article proposes a joint coding strategy to perform the coding of high-density stripes. More specifically, codewords are simultaneously embedded into the phase domain and the intensity domain of composite fringe patterns, which greatly extends the encoding range of valid codewords. Moreover, to improve the anti-interference capability of the proposed coding strategy, a regional correction technique (RCT) is presented to remove jump phase errors induced by the inevitable interference. The results of experiments indicate that the proposed method is effective and accurate for both complex objects and a standard sphere. [ABSTRACT FROM AUTHOR]

Natural language inference (NLI) task requires an agent to determine the semantic relation between a premise sentence (${p}$) and a hypothesis sentence (${h}$), which demands sufficient understanding about sentences semantic. Due to the issues, such as polysemy, ambiguity, as well as fuzziness of sentences, intense sentence understanding is very challenging. To this end, in this article, we introduce the corresponding image of sentences as reference information for enhancing sentence semantic understanding and representing. Specifically, we first propose an image-enhanced multilevel sentence representation net (IEMLRN), that utilizes the image features from pretrained models for enhancing the sentence semantic understanding at different scales, i.e., lexical-level, phrase-level, and sentence-level. The proposed model advances the performance on NLI tasks by leveraging the pretrained global features of images. However, as these pretrained image features are optimized on specific image classification datasets, they may not show the best performance on NLI tasks. Therefore, we further propose to design an adaptive image feature generator that extracts fine-grained image labels from the corresponding sentences. After that, we extend the IEMLRN to multilevel image-enhanced sentence representation net (MIESR) by utilizing not only the coarse-grained pretrained features of images, but also the fine-grained adaptive features of images. Therefore, sentence semantic can be evaluated and enhanced more comprehensively and precisely. Extensive experiments on two benchmark datasets (SNLI, SICK) clearly show our proposed IEMLRN significantly outperform the state-of-the-art baselines, and our proposed MIESR model achieves the best performance by considering not only the text but also images in an adaptive multigranularities way. [ABSTRACT FROM AUTHOR]

Multiple kernel learning (MKL), as a principled classification method, selects and combines base kernels to increase the categorization accuracy of Support Vector Machines (SVMs). The group method of data handling neural network (GMDH-NN) has been applied in many fields of optimization, data mining, and pattern recognition. It can automatically seek interrelatedness in data, select an optimal structure for the model or network, and enhance the accuracy of existing algorithms. We can utilize the advantages of the GMDH-NN to build a multiple graph kernel learning (MGKL) method and enhance the categorization performance of graph kernel SVMs. In this paper, we first define a unitized symmetric regularity criterion (USRC) to improve the symmetric regularity criterion of GMDH-NN. Second, a novel structure for the initial model of the GMDH-NN is defined, which uses the posterior probability output of graph kernel SVMs. We then use a hybrid graph kernel in the H 1 -space for MGKL in combination with the GMDH-NN. This way, we can obtain a pool of optimal graph kernels with different kernel parameters. Our experiments on standard graph datasets show that this new MGKL method is highly effective. • We propose a unitized symmetric regularity criterion. • We define a novel structure of initial model of GMDH-NN. • We use graph kernel SVM as base learner and integrate it with GMDH. • We design a special MGKL approach based on the GMDH-NN. • We use several scoring sorting methodologies to evaluate our experiments. [ABSTRACT FROM AUTHOR]

For offering proactive services (e.g., personalized exercise recommendation) to the students in computer supported intelligent education, one of the fundamental tasks is predicting student performance (e.g., scores) on future exercises, where it is necessary to track the change of each student's knowledge acquisition during her exercising activities. Unfortunately, to the best of our knowledge, existing approaches can only exploit the exercising records of students, and the problem of extracting rich information existed in the materials (e.g., knowledge concepts, exercise content) of exercises to achieve both more precise prediction of student performance and more interpretable analysis of knowledge acquisition remains underexplored. To this end, in this paper, we present a holistic study of student performance prediction. To directly achieve the primary goal of performance prediction, we first propose a general Exercise-Enhanced Recurrent Neural Network (EERNN) framework by exploring both student's exercising records and the text content of corresponding exercises. In EERNN, we simply summarize each student's state into an integrated vector and trace it with a recurrent neural network, where we design a bidirectional LSTM to learn the encoding of each exercise from its content. For making final predictions, we design two implementations on the basis of EERNN with different prediction strategies, i.e., EERNNM with Markov property and EERNNA with Attention mechanism. Then, to explicitly track student's knowledge acquisition on multiple knowledge concepts, we extend EERNN to an explainable Exercise-aware Knowledge Tracing (EKT) framework by incorporating the knowledge concept information, where the student's integrated state vector is now extended to a knowledge state matrix. In EKT, we further develop a memory network for quantifying how much each exercise can affect the mastery of students on multiple knowledge concepts during the exercising process. Finally, we conduct extensive experiments and evaluate both EERNN and EKT frameworks on a large-scale real-world data. The results in both general and cold-start scenarios clearly demonstrate the effectiveness of two frameworks in student performance prediction as well as the superior interpretability of EKT. [ABSTRACT FROM AUTHOR]

Location-based services (LBS) have become an important part of people's daily life. However, while providing great convenience for mobile users, LBS result in a serious problem on personal privacy, i.e., location privacy and query privacy. However, existing privacy methods for LBS generally take into consideration only location privacy or query privacy, without considering the problem of protecting both of them simultaneously. In this paper, we propose to construct a group of dummy query sequences, to cover up the query locations and query attributes of mobile users and thus protect users' privacy in LBS. First, we present a client-based framework for user privacy protection in LBS, which requires not only no change to the existing LBS algorithm on the server-side, but also no compromise to the accuracy of a LBS query. Second, based on the framework, we introduce a privacy model to formulate the constraints that ideal dummy query sequences should satisfy: (1) the similarity of feature distribution, which measures the effectiveness of the dummy query sequences to hide a true user query sequence; and (2) the exposure degree of user privacy, which measures the effectiveness of the dummy query sequences to cover up the location privacy and query privacy of a mobile user. Finally, we present an implementation algorithm to well meet the privacy model. Besides, both theoretical analysis and experimental evaluation demonstrate the effectiveness of our proposed approach, which show that the location privacy and attribute privacy behind LBS queries can be effectively protected by the dummy queries generated by our approach. [ABSTRACT FROM AUTHOR]