Search

Job-resume matching (JRM) is the core of online recruitment services for predicting the matching degree between a job post and a resume. Most of the existing methods for JRM achieve a promising performance by simplifying this task as a matching between the free-text attributes in the job post and the resume. However, they neglect the contributions of the semistructured multivariate attributes such as education and salary, which will result in an unsuccessful prediction. To address this issue, we propose a novel approach to comprehensively explore the Internal and EXternal InTeractions for semistructured multivariate attributes in JRM, i.e., InEXIT. In detail, we first encode the key and the value of each attribute as well as its source into the same semantic space. Next, to explore the complex relationships among the multivariate attributes, we propose to hierarchically model the internal interactions among the multivariate attributes inside the job post and the resume, as well as the external interactions between the job post and the resume. In particular, a stepwise fusion mechanism is designed to respectively integrate the key embeddings and the source embeddings into the value embeddings so as to clearly indicate the key and the source of the value. Finally, we employ an aggregation matching layer to predict the matching degree. We quantify the improvements of InEXIT against the competitive baselines on a real-world dataset, showing a general improvement of 4.28%, 4.10%, and 3.56% over the state-of-the-art baseline in terms of AUC, accuracy, and F1 score, respectively.

Generating semantically and emotionally context-consistent responses is key to intelligent dialogue systems. Previous works mainly refer to the context in the dialogue history to generate semantically related responses, ignoring the potential emotion in the conversation. In addition, existing methods mainly fail to consider the emotional changes of interlocutors and emotional categories simultaneously. However, emotion is crucial to reflect the interlocutor’s intent. In this paper, we propose an Emotion Capture Chat Machine (ECCM) that is able to capture the explicit and underlying emotional signal in the context to generate appropriate responses. In detail, we design a hierarchical recursive encoder-decoder framework with two enhanced self-attention encoders to capture the semantic signal and emotional signal, respectively, which are then fused in the decoder to produce the response. In general, we consider the dynamic and potential information of emotion to generate the response in multi-turn dialogues in the field of both daily conversation and psychological counseling. Our experimental results on a daily Chinese conversation dataset and a psychological counseling dataset show that ECCM outperforms the state-of-the-art baselines in terms of Perplexity, Distinct-1, Distinct-2, and manual evaluation. In addition, we find that ECCM performs well for input contexts with different lengths.

Event representation aims to transform individual events from a narrative event chain into a set of low-dimensional vectors to help support a series of downstream applications, e.g., similarity differentiation and missing event prediction. Traditional event representation models tend to focus on single modeling perspectives and thus are incapable of capturing physically disconnected yet semantically connected event segments. We, therefore, propose a heterogeneous event graph model (HeterEvent) to explicitly represent such event segments. Furthermore, another challenge in traditional event representation models is inherited from the datasets themselves. Data sparsity and insufficient labeled data are commonly encountered in event chains, easily leading to overfitting and undertraining. Therefore, we extend HeterEvent with a multistructure contrastive learning framework (MulCL) to alleviate the training risks from two structural perspectives. From the sequential perspective, a sequential-view contrastive learning component (SeqCL) is designed to facilitate the acquisition of sequential characteristics. From the graph perspective, a graph-view contrastive learning component (GraCL) is proposed to enhance the robustness of graph training by comparing different corrupted graphs. Experimental results confirm that our proposed MulCL

Session-aware recommendation is a special form of sequential recommendation, where users' previous interactions before current session are available. Recently, Recurrent Neural Network (RNN) based models are widely used in sequential recommendation tasks with great success. Previous works mainly focus on the interaction sequences of the current session without analyzing a user's long-term preferences. In this paper, we propose a joint neural network (JNN) for session-aware recommendation, which employs a Convolutional Neural Network(CNN) and a RNN to process the long-term historical interactions and the short-term sequential interactions respectively. Then, we apply a fully-connected neural network to study the complex relationship between these two types of features, which aims to generate a unified representation of the current session. Finally, a recommendation score for given items is generated by a bi-linear scheme upon the session representation. We conduct our experiments on three public datasets, showing that JNN outperforms the state-of-the-art baselines on all datasets in terms of Recall and Mean Reciprocal Rank (MRR). The outperforming results indicate that proper handling of historical interactions can improve the effectiveness of recommendation. The experimental results show that JNN is more prominent in samples with short current session or long historical interactions.

Hierarchical context modeling plays an important role in the response generation for multi-turn conversational systems. Previous methods mainly model context as multiple independent utterances and rely on attention mechanisms to obtain the context representation. They tend to ignore the explicit responds-to relationships between adjacent utterances and the special role that the user's latest utterance (the query) plays in determining the success of a conversation. To deal with this, we propose a multi-turn response generation model named KS-CQ, which contains two crucial components, the Keep and the Select modules, to produce a neighbor-aware context representation and a context-enriched query representation. The Keep module recodes each utterance of context by attentively introducing semantics from its prior and posterior neighboring utterances. The Select module treats the context as background information and selectively uses it to enrich the query representing process. Extensive experiments on two benchmark multi-turn conversation datasets demonstrate the effectiveness of our proposal compared with the state-of-the-art baselines in terms of both automatic and human evaluations.

Collaborative filtering (CF) aims to make recommendations for users by detecting user’s preference from the historical user–item interactions. Existing graph neural networks (GNN) based methods achieve satisfactory performance by exploiting the high-order connectivity between users and items, however they suffer from the poor training efficiency problem and easily introduce bias for information propagation. Moreover, the widely applied Bayesian personalized ranking (BPR) loss is insufficient to provide supervision signals for training due to the extremely sparse observed interactions. To deal with the above issues, we propose the Efficient Graph Collaborative Filtering (EGCF) method. Specifically, EGCF adopts merely one-layer graph convolution to model the collaborative signal for users and items from the first-order neighbors in the user–item interactions. Moreover, we introduce contrastive learning to enhance the representation learning of users and items by deriving the self-supervisions, which is jointly trained with the supervised learning. Extensive experiments are conducted on two benchmark datasets, i.e., Yelp2018 and Amazon-book, and the experimental results demonstrate that EGCF can achieve the state-of-the-art performance in terms of Recall and normalized discounted cumulative gain (NDCG), especially on ranking the target items at right positions. In addition, EGCF shows obvious advantages in the training efficiency compared with the competitive baselines, making it practicable for potential applications.

Text representation, which converts text spans into real-valued vectors or matrices, is a crucial tool for machines to understand the semantics of text. Although most previous works employed classic methods based on statistics and neural networks, such methods might suffer from data sparsity and insensitivity to the text structure, respectively. To address the above drawbacks, we propose a general and structure-sensitive framework, i.e., the hierarchical architecture. Specifically, we incorporate the hierarchical architecture into three existing neural network models for document representation, thereby producing three new representation models for document classification, i.e., TextHFT, TextHRNN, and TextHCNN. Our comprehensive experimental results on two public datasets demonstrate the effectiveness of the hierarchical architecture. With a comparable (or substantially less) time expense, our proposals obtain significant improvements ranging from 4.65 to 35.08% in terms of accuracy against the baseline. We can conclude that the hierarchical architecture can enhance the classification performance. In addition, we find that the benefits provided by the hierarchical architecture can be strengthened as the document length increases.

Answer selection is an essential step in a question answering (QA) system. Traditional methods for this task mainly focus on developing linguistic features that are limited in practice. With the great success of deep learning method in distributed text representation, deep learning-based answer selection approaches have been well investigated, which mainly employ only one neural network, i.e., convolutional neural network (CNN) or long short term memory (LSTM), leading to failures in extracting some rich sentence features. Thus, in this paper, we propose a collaborative learning-based answer selection model (QA-CL), where we deploy a parallel training architecture to collaboratively learn the initial word vector matrix of the sentence by CNN and bidirectional LSTM (BiLSTM) at the same time. In addition, we extend our model by incorporating the sentence embedding generated by the QA-CL model into a joint distributed sentence representation using a strong unsupervised baseline weight removal (WR), i.e., the QA-CLWR model. We evaluate our proposals on a popular QA dataset, InsuranceQA. The experimental results indicate that our proposed answer selection methods can produce a better performance compared with several strong baselines. Finally, we investigate the models’ performance with respect to different question types and find that question types with a medium number of questions have a better and more stable performance than those types with too large or too small number of questions.

Answer selection is a crucial subtask in the question answering (QA) system. Conventional avenues for this task mainly concentrate on developing linguistic tools that are limited in both performance and practicability. Answer selection approaches based on deep learning have been well investigated with the tremendous success of deep learning in natural language processing. However, the traditional neural networks employed in existing answer selection models, i.e., recursive neural network or convolutional neural network, typically suffer from obtaining the global text information due to their operating mechanisms. The recent Transformer neural network is considered to be good at extracting the global information by employing only self-attention mechanism. Thus, in this paper, we design a Transformer-based neural network for answer selection, where we deploy a bidirectional long short-term memory (BiLSTM) behind the Transformer to acquire both global information and sequential features in the question or answer sentence. Different from the original Transformer, our Transformer-based network focuses on sentence embedding rather than the seq2seq task. In addition, we employ a BiLSTM rather than utilizing the position encoding to incorporate sequential features as the universal Transformer does. Furthermore, we apply three aggregated strategies to generate sentence embeddings for question and answer, i.e., the weighted mean pooling, the max pooling, and the attentive pooling, leading to three corresponding Transformer-based models, i.e., QA-TF $_{{W\!P}}$ , QA-TF $_{{M\!P}}$ , and QA-TF $_{{A\!P}}$ , respectively. Finally, we evaluate our proposals on a popular QA dataset WikiQA. The experimental results demonstrate that our proposed Transformer-based answer selection models can produce a better performance compared with several competitive baselines. In detail, our best model outperforms the state-of-the-art baseline by up to 2.37%, 2.83%, and 3.79% in terms of MAP, MRR, and accuracy, respectively.

Session-based recommendation aims to model a user’s intent and predict an item that the user may interact with in the next step based on an ongoing session. Existing session-based recommender systems mainly aim to model the sequential signals based on Recurrent Neural Network (RNN) structures or the item transition relations between items with Graph Neural Network (GNN) based frameworks to identify a user’s intent for recommendation. However, in real scenarios, there may be strong sequential signals existing in users’ adjacent behaviors or multi-step transition relations among different items. Thus, either RNN- or GNN-based methods can only capture limited information for modeling complex user behavior patterns. RNNs pay attention to the sequential relations among consecutive items, while GNNs focus on structural information, i.e., how to enrich the item embedding with its adjacent items. In this paper, we propose a Collaborative Co-attention Network for Session-based Recommendation (CCN-SR) to incorporate both sequential and structural information, as well as capture the co-relations between them for obtaining an accurate session representation. To be specific, we first model the ongoing session with an RNN structure to capture the sequential information among items. Meanwhile, we also construct a session graph to learn the item representations with a GNN structure. Then, we design a co-attention network upon these two structures to capture the mutual information between them. The designed co-attention network can enrich the representation of each node in the session with both sequential and structural information, and thus generate a more comprehensive representation for each session. Extensive experiments are conducted on two public e-commerce datasets, and the results demonstrate that our proposed model outperforms state-of-the-art baseline model for session based recommendation in terms of both Recall and MRR. We also investigate different combination strategies and the experimental results verify the effectiveness of our proposed co-attention mechanism. Besides, our CCN-SR model achieves better performance than baseline models with different session lengths.

Event detection classifies unlabeled sentences into event labels, which can benefit numerous applications, including information retrieval, question answering and script learning. One of the major obstacles to event detection in reality is insufficient training data. To deal with the low-resources problem, we investigate few-shot event detection in this paper and propose TaLeM, a novel taxonomy-aware learning model, consisting of two components, i.e., the taxonomy-aware self-supervised learning framework (TaSeLF) and the taxonomy-aware prototypical networks (TaPN). Specifically, TaSeLF mines the taxonomy-aware distance relations to increases the training examples, which alleviates the generalization bottleneck brought by the insufficient data. TaPN introduces the Poincare embeddings to represent the label taxonomy, and integrates them into a task-adaptive projection networks, which tackles problems of the class centroids distribution and the taxonomy-aware embedding distribution in the vanilla prototypical networks. Extensive experiments in the four types of meta tasks demonstrate the superiority of our proposal over the strong baselines, and further verify the effectiveness and importance of modeling the label taxonomy. Besides, TaSeLF can be a flexible plug-in for the other taxonomy-based few-shot classification tasks.

Text representation can aid machines in understanding text. Previous work on text representation often focuses on the so-called forward implication, i.e., preceding words are taken as the context of later words for creating representations, thus ignoring the fact that the semantics of a text segment is a product of the mutual implication of words in the text: later words contribute to the meaning of preceding words. We introduce the concept of interaction and propose a two-perspective interaction representation, that encapsulates a local and a global interaction representation. Here, a local interaction representation is one that interacts among words with parent-children relationships on the syntactic trees and a global interaction interpretation is one that interacts among all the words in a sentence. We combine the two interaction representations to develop a Hybrid Interaction Representation (HIR). Inspired by existing feature-based and fine-tuning-based pretrain-finetuning approaches to language models, we integrate the advantages of feature-based and fine-tuning-based methods to propose the Pre-train, Interact, Fine-tune (PIF) architecture. We evaluate our proposed models on five widely-used datasets for text classification tasks. Our ensemble method, outperforms state-of-the-art baselines with improvements ranging from 2.03% to 3.15% in terms of error rate. In addition, we find that, the improvements of PIF against most state-of-the-art methods is not affected by increasing of the length of the text., 32 pages, 5 figures

Session-based recommendation is a challenging task. Without access to a user's historical user-item interactions, the information available in an ongoing session may be very limited. Previous work on session-based recommendation has considered sequences of items that users have interacted with sequentially. Such item sequences may not fully capture complex transition relationship between items that go beyond inspection order. Thus graph neural network (GNN) based models have been proposed to capture the transition relationship between items. However, GNNs typically propagate information from adjacent items only, thus neglecting information from items without direct connections. Importantly, GNN-based approaches often face serious overfitting problems. We propose Star Graph Neural Networks with Highway Networks (SGNN-HN) for session-based recommendation. The proposed SGNN-HN applies a star graph neural network (SGNN) to model the complex transition relationship between items in an ongoing session. To avoid overfitting, we employ highway networks (HN) to adaptively select embeddings from item representations. Finally, we aggregate the item embeddings generated by the SGNN in an ongoing session to represent a user's final preference for item prediction. Experiments on two public benchmark datasets show that SGNN-HN can outperform state-of-the-art models in terms of P@20 and MRR@20 for session-based recommendation.

Given an occurred event, human can easily predict the next event or reason the preceding event, yet which is difficult for machine to perform such event reasoning. Event representation bridges the connection and targets to model the process of event reasoning as a machine-readable format, which then can support a wide range of applications in information retrieval, e.g., question answering and information extraction. Existing work mainly resorts to a joint training to integrate all levels of training loss in event chains by a simple loss summation, which is easily trapped into a local optimum. In addition, the scenario knowledge in event chains is not well investigated for event representation. In this paper, we propose a unified fine-tuning architecture, incorporated with scenario knowledge for event representation, i.e., UniFA-S, which mainly consists of a unified fine-tuning architecture (UniFA) and a scenario-level variational auto-encoder (S-VAE). In detail, UniFA employs a multi-step fine-tuning to integrate all levels of training and S-VAE applies a stochastic variable to implicitly represent the scenario-level knowledge. We evaluate our proposal from two aspects, i.e., the representation and inference abilities. For the representation ability, our ensemble model UniFA-S can beat state-of-the-art baselines for two similarity tasks. For the inference ability, UniFA-S can outperform the best baseline, achieving 4.1%-8.2% improvements in terms of accuracy for various inference tasks.

The application of pose assessment on rehabilitation training has gradually received attention in recent years. However, current evaluation indicators of these methods are mostly based on the score or scoring function that defined by users, which is too subjective and hard to be used by patients directly. In this paper, we conceptualized a new idea for pose matching, namely pose-guided matching that aims at providing objective and accurate score, feedback and guidance (i.e. guided) to the patients when the pose is compared to the standard pose. More specifically, we proposed a pair-based Siamese Convolutional Neural Network (SCNN) abbreviated ST-AMCNN to realize the idea of pose-guided matching on the eight-section brocade dataset which is one of the most representative traditional rehabilitation exercises in China. We simplified the multi-stages pose matching by merging two standalone modules (i.e. alignment and matching module) into a one-stage task. Such that, only one loss function is required to tune, which reduces the computational complexity. On top of the Spatial Transformer Networks (STN) employed as an alignment module, we proposed a new Attention-based Multi-Scale Convolution (AMC) to match different posture parts (i.e. multi-scale). Furthermore, the proposed AMC can assign more weight to useful pose features as opposed to other irrelevant features e.g. background features for performance gain. Finally, Gradient-weighted Class Activation Mapping (Grad-CAM) is adopted to visualize the matching result for the learner. Experimental results indicate that ST-AMCNN achieves a competitive performance than the state-of-the-art models and can provide accurate feedback for learners on rehabilitation training. Simultaneously, the proposed method is also deployed in client software for testing.

Fact verification aims to retrieve related evidence from raw text to verify the correctness of a given claim. Existing works mainly leverage the single-granularity features for the representation learning of evidences, i.e., sentence features, ignoring other features like entity-level and context-level features. In addition, they usually focus on improving the prediction accuracy while lacking the interpretability of the inference process, which leads to unreliable results. Thus, in this paper, to investigate how to utilize multi-granularity semantic units for evidence representation as well as to improve the explainability of evidence reasoning, we propose a Hierarchical Reasoning-based Heterogeneous Graph Neural Network for fact verification (HHGN). HHGN combines multiple features of entity, sentence as well as context for evidence representation, and employs a heterogeneous graph to capture their semantic relations. Inspired by the human inference process, we design a hierarchical reasoning-based node updating strategy to propagate the evidence features. Then, we extract the potential reasoning paths from the graph to predict the label, which aggregates the results of different paths weighted by their relevance to the claim. We evaluate our proposal on FEVER, a large-scale benchmark dataset for fact verification. Our experimental results demonstrate the superiority of HHGN over the competitive baselines in both single evidence and multiple evidences settings. In addition, HHGN presents reasonable interpretability in the form of aggregating the features of relevant entity units and selecting the evidence sentences with high confidence.

Fact verification aims to evaluate the authenticity of a given claim based on the evidence sentences retrieved from Wikipedia articles. Existing works mainly leverage the natural language inference methods to model the semantic interaction of claim and evidence, or further employ the graph structure to capture the relation features between multiple evidences. However, previous methods have limited representation ability in encoding complicated units of claim and evidences, and thus cannot support sophisticated reasoning. In addition, a limited amount of supervisory signals lead to the graph encoder could not distinguish the distinctions of different graph structures and weaken the encoding ability. To address the above issues, we propose a Knowledge-Enhanced Graph Attention network (KEGA) for fact verification, which introduces a knowledge integration module to enhance the representation of claims and evidences by incorporating external knowledge. Moreover, KEGA leverages an auxiliary loss based on contrastive learning to fine-tune the graph attention encoder and learn the discriminative features for the evidence graph. Comprehensive experiments conducted on FEVER, a large-scale benchmark dataset for fact verification, demonstrate the superiority of our proposal in both the multi-evidences and single-evidence scenarios. In addition, our findings show that the background knowledge for words can effectively improve the model performance.

We propose a J-NCF method for recommender systems. The J-NCF model applies a joint neural network that couples deep feature learning and deep interaction modeling with a rating matrix. Deep feature learning extracts feature representations of users and items with a deep learning architecture based on a user-item rating matrix. Deep interaction modeling captures non-linear user-item interactions with a deep neural network using the feature representations generated by the deep feature learning process as input. J-NCF enables the deep feature learning and deep interaction modeling processes to optimize each other through joint training, which leads to improved recommendation performance. In addition, we design a new loss function for optimization, which takes both implicit and explicit feedback, point-wise and pair-wise loss into account. Experiments on several real-word datasets show significant improvements of J-NCF over state-of-the-art methods, with improvements of up to 8.24% on the MovieLens 100K dataset, 10.81% on the MovieLens 1M dataset, and 10.21% on the Amazon Movies dataset in terms of HR@10. NDCG@10 improvements are 12.42%, 14.24% and 15.06%, respectively. We also conduct experiments to evaluate the scalability and sensitivity of J-NCF. Our experiments show that the J-NCF model has a competitive recommendation performance with inactive users and different degrees of data sparsity when compared to state-of-the-art baselines., 30 pages

Answer selection focuses on selecting the correct answer for a question. Most previous work on answer selection achieves good performance by employing an RNN, which processes all question and answer sentences with the same feature extractor regardless of the sentence length. These methods often encounter the problem of long-term dependencies. To address this issue, we propose a Length-adaptive Neural Network (LaNN) for answer selection that can auto-select a neural feature extractor according to the length of the input sentence. In particular, we propose a flexible neural structure that applies a BiLSTM-based feature extractor for short sentences and a Transformer-based feature extractor for long sentences. To the best of our knowledge, LaNN is the first neural network structure that can auto-select the feature extraction mechanism based on the input. We quantify the improvements of LaNN against several competitive baselines on the public WikiQA dataset, showing significant improvements over the state-of-the-art.

Fact verification aims to retrieve relevant evidence from a knowledge base, e.g., Wikipedia, to verify the given claims. Existing methods only consider the sentence-level semantics for evidence representations, which typically neglect the importance of fine-grained features in the evidence-related sentences. In addition, the interpretability of the reasoning process has not been well studied in the field of fact verification. To address such issues, we propose an entity-graph based reasoning method for fact verification abbreviated as RoEG, which generates the fine-grained features of evidence at the entity-level and models the human reasoning paths based on an entity graph. In detail, to capture the semantic relations of retrieved evidence, RoEG introduces the entities as nodes and constructs the edges in the graph based on three linking strategies. Then, RoEG utilizes a selection gate to constrain the information propagation in the sub-graph of relevant entities and applies a graph neural network to propagate the entity-features for reasoning. Finally, RoEG employs an attention aggregator to gather the information of entities for label prediction. Experimental results on a large-scale benchmark dataset FEVER demonstrate the effectiveness of our proposal by beating the competitive baselines in terms of label accuracy and FEVER Score. In particular, for a task of multiple-evidence fact verification, RoEG produces 5.48% and 4.35% improvements in terms of label accuracy and FEVER Score against the state-of-the-art baseline. In addition, RoEG shows a better performance when more entities are involved for fact verification.

Incorporating topic information can help response generation models to produce informative responses for chat-bots. Previous work only considers the individual semantic of each topic, ignoring its specific dialog context, which may result in inaccurate topic representation and hurt response coherence. Besides, as an important feature of multi-turn conversation, dynamic topic transitions have not been well-studied. We propose a Context-Controlled Topic-Aware neural response generation model, i.e., CCTA, which makes dialog context interact with the process of topic representing and transiting to achieve balanced improvements on response informativeness and contextual coherence. CCTA focuses on capturing the semantical relations within topics as well as their corresponding contextual information in conversation, to produce context-dependent topic representations at the word-level and turn-level. Besides, CCTA introduces a context-controlled topic transition strategy, utilizing contextual topics to yield relevant transition words. Extensive experimental results on two benchmark multi-turn conversation datasets validate the superiority of our proposal on generating coherent and informative responses against the state-of-the-art baselines. We also find that topic transition modeling can work as an auxiliary learning task to boost the response generation.

In previous work on text summarization, encoder-decoder architectures and attention mechanisms have both been widely used. Attention-based encoder-decoder approaches typically focus on taking the sentences preceding a given sentence in a document into account for document representation, failing to capture the relationships between a sentence and sentences that follow it in a document in the encoder. We propose an attentive encoder-based summarization (AES) model to generate article summaries. AES can generate a rich document representation by considering both the global information of a document and the relationships of sentences in the document. A unidirectional recurrent neural network (RNN) and a bidirectional RNN are considered to construct the encoders, giving rise to unidirectional attentive encoder-based summarization (Uni-AES) and bidirectional attentive encoder-based summarization (Bi-AES), respectively. Our experimental results show that Bi-AES outperforms Uni-AES. We obtain substantial improvements over a relevant start-of-the-art baseline.

Query auto-completion (QAC) is one of the most visible features in modern search engines. It helps users complete their queries by presenting a list of possible completions while they are typing in the search box. Existing works on QAC focus on employing learning-to-rank algorithms over handcrafted features. However, those manually designed features are unable to capture non-linear relationships between users and their submitted queries. Meanwhile, although Recurrent Neural Networks (RNNs) show significant advances in various areas, little attention is paid to its application to QAC. To bridge this gap, we propose three RNN-based models for QAC ranking: a simple session-based RNN model, a personalized RNN model and an attentive RNN model. Extensive experiments are conducted on a real-world query log. The significant improvement over the compared baseline verifies the effectiveness of the personalized RNN model and the attentive RNN model.

Users are strongly influenced by their friends or other users with similar interest. In this paper, we first provide a friends cluster identification approach based on the analysis of social features. Secondly, we propose our static recommendation approach based on an unbiased random walk strategy which simultaneously considers traditional recommendation approach and social relationship. Then, we further identify the change of both friendship and interest over time and propose our extended recommendation algorithm. Finally, we evaluate our approach on CiteULike and Last.fm dataset. Our experimental results demonstrate that the proposed algorithms can be very effective in recommending unknown items.

Document classification has a broad application in the field of sentiment classification, document ranking and topic labeling, etc. Previous neural network-based work has mainly focused on investigating a so-called forward implication, i.e., the preceding text segments are taken as the context of the following text segments when generating the text representation. Such a scenario typically ignores the fact that the semantics of a document are a product of the mutual implication of all text segments in a document. Thus, in this paper, we introduce a concept of interaction and propose a text representation model with Self-interaction Attention Mechanism (TextSAM) for document classification. In particular, we design three aggregated strategies to integrate the interaction into a hierarchical architecture for document classification, i.e., averaging the interaction, maximizing the interaction and adding one more attention layer on the interaction, which leads to three models, i.e., TextSAMAVE, TextSAMMAX and TextSAMATT, respectively. Our comprehensive experimental results on two public datasets, i.e., Yelp 2016 and Amazon Reviews (Electronics), show that our proposals can significantly outperform the state-of-the-art neural-based baselines for document classification, presenting a general improvement in terms of accuracy ranging from 5.97% to 14.05% against the best baseline. Furthermore, we find that our proposals with a self-interaction attention mechanism can obviously alleviate the impact brought by the increase of sentence number as the relative improvement of our proposals against the baselines are enlarged when the sentence number increases.

Document representation is widely used in practical application, for example, sentiment classification, text retrieval, and text classification. Previous work is mainly based on the statistics and the neural networks, which suffer from data sparsity and model interpretability, respectively. In this paper, we propose a general framework for document representation with a hierarchical architecture. In particular, we incorporate the hierarchical architecture into three traditional neural-network models for document representation, resulting in three hierarchical neural representation models for document classification, that is, TextHFT, TextHRNN, and TextHCNN. Our comprehensive experimental results on two public datasets, that is, Yelp 2016 and Amazon Reviews (Electronics), show that our proposals with hierarchical architecture outperform the corresponding neural-network models for document classification, resulting in a significant improvement ranging from 4.65% to 35.08% in terms of accuracy with a comparable (or substantially less) expense of time consumption. In addition, we find that the long documents benefit more from the hierarchical architecture than the short ones as the improvement in terms of accuracy on long documents is greater than that on short documents.

创新点 本文提出一种贪婪算法, 解决信息检索系统中查询推荐多样化问题, 算法目的旨在返回给用户的查询推荐列表既能准确包含用户的潜在查询, 又能使得查询列表涵盖尽可能多的主题, 这样提高不同类型用户查询推荐满意度。 在本算法中, 用户的查询意图不仅体现在当前查询热度上, 同时我们从用户的检索查询历史中挖掘有用信息预测用户意图, 生成用户的查询意图在各个主题上的概率分布, 并依此计算每个查询词被提交的概率并进行排序。 提出的算法在公共测试集上取得了较好地性能, 能把初始查询推荐列表中相似的查询词移除, 达到查询词多样化的目的。

Automatic aerial image interpretation is one of new rising high-tech application fields, and it’s proverbially applied in the military domain. Based on human visual attention mechanism and texture visual perception, this paper proposes a new approach for man-made object detection and marking by extracting texture and geometry structure features. After clustering the texture feature to realize effective image segmentation, geometry structure feature is obtained to achieve final detection and marking. Thus a man-made object detection methodology is designed, by which typical man-made objects in complex natural background, including airplanes, tanks and vehicles can be detected. The experiments sustain that the proposed method is effective and rational.

A fundamental challenge in image engineering is how to locate interested objects from high-resolution images with efficient detection performance. Several man-made objects detection approaches have been proposed while the majority of these methods are not truly timesaving and suffer low degree of detection precision. To address this issue, we propose a novel approach for man-made object detection in aerial image involving MapReduce scheme for large scale image analysis to support image feature extraction, which can be widely used to compute-intensive tasks in a highly parallel way, and texture feature extraction and clustering. Comprehensive experiments show that the parallel framework saves voluminous time for feature extraction with satisfied objects detection performance.

Information retrieval applications have to publish their output in the form of ranked lists. Such a requirement motivates researchers to develop methods that can automatically learn effective ranking models. Many existing methods usually perform analysis on multidimensional features of query-document pairs directly and don't take users' interactive feedback information into account. They thus incur the high computation overhead and low retrieval performance due to an indefinite query expression. In this paper, we propose a Virtual Feature based Logistic Regression (VFLR) ranking method that conducts the logistic regression on a set of essential but independent variables, called virtual features (VF). They are extracted via the principal component analysis (PCA) method with the user's relevance feedback. We then predict the ranking score of each queried document to produce a ranked list. We systematically evaluate our method using the LETOR 4.0 benchmark datasets. The experimental results demonstrate that the proposal outperforms the state-of-the-art methods in terms of the Mean Average Precision (MAP), the Precision at position k (P@k), and the Normalized Discounted Cumulative Gain at position k (NDCG@k).