Recently, the prediction of drug-target interactions (DTIs) is improved by graph neural networks (GNNs). There exists some DTIs that are far away from other DTIs, which plays fundamental role in drug development. However, conventional GNNs encode long-range topological correlations with latent embeddings, restricting their performances in inferring distant DTIs. Inspired by random walk methods that directly encode the connectivity between nodes by transition probabilities, a random walk guided graph neural network (RWGNN) method is proposed, in which random walk profiles are passed through a GNN to enable it to learn distance-aware node embeddings. For an unknown DTI, enclosing subnetworks are firstly extracted. Node-level, interaction-level and network-level transition probabilities of random walks (i.e., random walk profile) are calculated on these subnetworks. Then, each graph convolutional layer aggregates node embeddings from multi-hop neighbors according to attentional weights calculated from random walk profiles. Finally, the DTI’s drug embedding, protein embedding and random walk profile are aggregated to calculate an interaction score for it. The performance in inferring distant DTIs (≥ 3 hops away from known DTIs) has been improved by RWGNN (AUC=0.957) significantly, compared with GCN (AUC=0.582) and GIN (AUC=0.724). Besides, RWGNN is powerful in inferring DTIs for cold-start drugs and target proteins. Top-10 scored DTIs of RWGNN can be verified in literatures.

Recently, with the in-depth study of biological network structures, methods such as graph neural networks and graph contrastive learning have attracted significant attention and demonstrated notable advantages in DTI prediction. Nevertheless, it remains a challenging task that predicting new DTIs by a small number of known data in heterogeneous biological networks. Graph contrastive learning as an effective method to solute this issue has been proposed recently. Although contrastive learning have shown significant advantages, the objective function heavily relies on unbiased positive and negative samples. Inspired by this issue, a heterogeneous ranking contrastive learning method (HRCL-DTI) for DTI prediction is proposed. Specifically, multiple graph encoders are employed to capture the topological relationships in heterogeneous biological networks. After that, the prediction scores are calculated in the ranking module of HRCL-DTI to select reliable positive and negative samples for the heterogeneous contrastive learning module, aiming to enhance the consistency of DTI representations. Experimental results demonstrate that HRCL-DTI outperforms existing stateof-the-art baselines on multiple datasets and it possesses strong generalization ability and practical effectiveness.

Prioritizing casual genes for disease phenotypes can reveal the inherent basis of human diseases and facilitate drug developments. It is widely believed that the full utilization of hierarchical structure of phenotype ontologies in HPO can improve the performance of gene prioritization significantly. However, previous works mostly utilize the hierarchical structure as graph constraint or implicit hierarchical relation, thus unable to fully depict the underlying correlations among adjacent levels of phenotype ontologies.To tackle the issue, we proposed a Multi-View Nonnegative Matrix Factorization method with Layer-Wise Explicit Hierarchical Constraint (HMNMF), which fully explores the “is-a” relationship between adjacent levels of phenotypes in HPO. Experimental results demonstrate the outstanding performance of HMNMF not only in the extraordinary global gene prioritization performance but also in the strong ability to prioritize candidate genes for totally new phenotypes.

Specifying an appropriate skill set for a job position is critical for talent recruitment. However, it is often more difficult than people think since it needs a great understanding of the role of the position, related technologies, and even the global situation of the job market. To this end, we propose to learn the mapping between the job position description and its required skills in a data-driven manner. This task is challenging due to the complex mapping relationships between job descriptions and skills, which is caused by complex influencing factors. In this paper, we propose a novel Multi-Graph Neural Network based Skill Prediction model (MGNSP) to make skill prediction by learning effective deep semantics matching of job positions and skills. Specifically, to capture the complex heterogeneous relations among the job positions, skills, and meta information, we devise a joint learning approach of graph neural networks for multiple information networks, which are J-Net, S-Net and JS-Net, respectively. After jointly learning complementary semantics of job positions and skills with three multi-layer graph neural networks from these information networks, the skills are predicted by learning to match their representations. Extensive experimental results on a real-world dataset validate the effectiveness of our model.

The interpretable charge prediction task is to predict the final charges according to the fact descriptions, at the same time generate the corresponding explanations. The task is often more difficult than people think, as the explanation and prediction are highly interdependent, which is ignored by existing methods. In this paper, we deal with the interpretable charge prediction task from the perspective of generating the prediction and explanation (court view) pair while considering their dependency. To this end, we propose a Joint Prediction and Generation Model, named JPGM, which includes a coarse-to-fine classifier and a keyword-aware generator. Specifically, firstly, the classifier predicts a group of charge labels given the fact description. Then, we select the most matchable keywords of every charge from the pre-defined charge-discriminative keywords by an attention mechanism. Furthermore, the generator generates explanation using both the fact description and the most matchable keywords. Finally, to refine the prediction, the generated explanation and fact description are fused for final charge prediction. The classifier and generator are trained under an alternating procedure, which alleviates the error propagation. The experimental results validate that our model can effectively address the dependency issue and predict the charge with more interpretability.

This paper concerns the problem of network embedding (NE), which aims to learn low-dimensional representations for network nodes. Such dense representations offer great promises for many network analysis problems. However, existing approaches are still faced with challenges posed by the characteristics of complex real-world networks. First, for networks associated with rich content information, previous methods often learn separated content and structure representations, which requires post-processing of combination. Empirical combination strategies often make the final vectors suboptimal. Second, existing methods preserve the structure information by considering short and fixed neighborhood scope, such as the first- and/or the second-order proximities. However, it is hard to decide the neighborhood scope in complex problems. To this end, we propose a novel sequence to sequence model based NE framework referred to as Self-Translation Network Embedding (STNE). With the sampled node sequences, STNE translates each sequence itself from the content sequence to the node sequence. On the one hand, the bi-directional LSTM encoder fuses the content and structure information seamlessly from the raw input. On the other hand, high-order proximity can be flexibly learned with the memories of LSTM to capture long-range structural information. Experimental results on three real-world datasets demonstrate the superiority of STNE.

Visual place recognition is a challenging problem in simultaneous localization and mapping (SLAM) due to a large variability of the scene appearance. A place is usually described by a single-frame image in conventional place recognition algorithms. However, it is unlikely to completely describe the place appearance using a single frame image. Moreover, it is more sensitive to the change of environments. In this article, a novel image-sequence-based framework for place detection and recognition is proposed. Rather than a single frame image, a place is represented by an image sequence in this article. Position invariant robust feature (PIRF) descriptors are extracted from images and processed by the incremental bag-of-words (BoWs) for feature extraction. The robot automatically partitions the sequentially acquired images into different image sequences according to the change of the environmental appearance. Then, the echo state network (ESN) is applied to model each image sequence. The resultant states of the ESN are used as features of the corresponding image sequence for place recognition. The proposed method is evaluated on two public datasets. Experimental comparisons with the FAB-MAP 2.0 and SeqSLAM are conducted. Finally, a real-world experiment on place recognition with a mobile robot is performed to further verify the proposed method.

Writing a good job posting is a critical step in the recruiting process, but the task is often more difficult than many people think.It is challenging to specify the level of education, experience, relevant skills per the company information and job description.To this end, we propose a novel task of Job Posting Generation (JPG) that is cast as a conditional text generation problem to generate job requirements according to the job descriptions.To deal with this task, we devise a data-driven global Skill-Aware Multi-Attention generation model, named SAMA.Specifically, to model the complex mapping relationships between input and output, we design a hierarchical decoder that we first label the job description with multiple skills, then we generate a complete text guided by the skill labels.At the same time, to exploit the prior knowledge about the skills, we further construct a skill knowledge graph to capture the global prior knowledge of skills and refine the generated results.The proposed approach is evaluated on real-world job posting data.Experimental results clearly demonstrate the effectiveness of the proposed method 1 .

Identification of new interactions between drugs and target proteins (DTIs) plays a fundamental role in drug development. It is commonly recognized that the collaborative utilization of drug-drug interaction (DDI) networks and protein-protein interaction (PPI) networks contribute to more comprehensive prediction results. However, recent methods almost view the different types of nodes and edges in heterogeneous networks indiscriminately, thus neglecting the complementary information hidden across different types of interactions. Therefore, this work innovatively elaborates a Multi-Resolutional Collaborative Heterogeneous Graph Convolutional Auto-Encoder (MRCH-GCAE) for DTI prediction, which collaboratively aggregates the learned embeddings from different types of links in heterogeneous drugtarget networks, thus leading to more interpretable embeddings for each drug and target node. Experiments have demonstrated the outstanding effectiveness of it not only in the prediction accuracy but also the ability to predict DTIs for new drugs and new target proteins.

Recently, network embedding (NE) has achieved great successes in learning low dimensional representations for network nodes and has been increasingly applied to various network analytic tasks. In this paper, we consider the representation learning problem for content-rich networks whose nodes are associated with rich content information. Content-rich network embedding is challenging in fusing the complex structural dependencies and the rich contents. To tackle the challenges, we propose a generative model, Network-to-Network Network Embedding (Net2Net-NE) model, which can effectively fuse the structure and content information into one continuous embedding vector for each node. Specifically, we regard the content-rich network as a pair of networks with different modalities, i.e., content network and node network. By exploiting the strong correlation between the focal node and the nodes to whom it is connected to, a multilayer recursively composable encoder is proposed to fuse the structure and content information of the entire ego network into the egocentric node embedding. Moreover, a cross-modal decoder is deployed to mapping the egocentric node embeddings into node identities in an interconnected network. By learning the identity of each node according to its content, the mapping from content network to node network is learned in a generative manner. Hence the latent encoding vectors learned by the Net2Net-NE can be used as effective node embeddings. Extensive experimental results on three real-world networks demonstrate the superiority of Net2Net-NE over state-of-the-art methods.

We study multi-turn response generation in chatbots where a response is generated according to a conversation context. Existing work has modeled the hierarchy of the context, but does not pay enough attention to the fact that words and utterances in the context are differentially important. As a result, they may lose important information in context and generate irrelevant responses. We propose a hierarchical recurrent attention network (HRAN) to model both the hierarchy and the importance variance in a unified framework. In HRAN, a hierarchical attention mechanism attends to important parts within and among utterances with word level attention and utterance level attention respectively.

Hashtags have always been important elements in many social network platforms and micro-blog services. Semantic understanding of hashtags is a critical and fundamental task for many applications on social networks, such as event analysis, theme discovery, information retrieval, etc. However, this task is challenging due to the sparsity, polysemy, and synonymy of hashtags. In this paper, we investigate the problem of hashtag embedding by combining the short text content with the various heterogeneous relations in social networks. Specifically, we first establish a network with hashtags as its nodes. Hierarchically, each of the hashtag nodes is associated with a set of tweets and each tweet contains a set of words. Then we devise an embedding model, called Hashtag2Vec, which exploits multiple relations of hashtag-hashtag, hashtag-tweet, tweet-word, and word-word relations based on the hierarchical heterogeneous network. In addition to embedding the hashtags, our proposed framework is capable of embedding the short social texts as well. Extensive experiments are conducted on two real-world datasets, and the results demonstrate the effectiveness of the proposed method.

This paper concerns the problem of network embedding (NE), whose aim is to learn low-dimensional representations for nodes in networks. Such dense vector representations offer great promises for many network analysis problems. However, existing NE approaches are still faced with challenges posed by the characteristics of complex networks in real-world applications. First, for many real-world networks associated with rich content information, previous NE methods tend to learn separated content and structure representations for each node, which requires a post-processing of combination. The empirical and simple combination strategies often make the final vector suboptimal. Second, the existing NE methods preserve the structure information by considering short and fixed neighborhood scope, such as the first- and/or the second-order proximities. However, it is hard to decide the scope of the neighborhood when facing a complex problem. To this end, we propose a novel sequence-to-sequence model based NE framework which is referred to as Self-Translation Network Embedding (STNE) model. With the sequences generated by random walks on a network, STNE learns the mapping that translates each sequence itself from the content sequence to the node sequence. On the one hand, the bi-directional LSTM encoder of STNE fuses the content and structure information seamlessly from the raw input. On the other hand, high-order proximity can be flexibly learned with the memories of LSTM to capture long-range structural information. By such self-translation from content to node, the learned hidden representations can be adopted as node embeddings. Extensive experimental results based on three real-world datasets demonstrate that the proposed STNE outperforms the state-of-the-art NE approaches. To facilitate reproduction and further study, we provide Internet access to the code and datasets\footnotehttp://dm.nankai.edu.cn/code/STNE.rar.

BACKGROUND: Prioritizing disease genes is trying to identify potential disease causing genes for a given phenotype, which can be applied to reveal the inherited basis of human diseases and facilitate drug development. Our motivation is inspired by label propagation algorithm and the false positive protein-protein interactions that exist in the dataset. To the best of our knowledge, the false positive protein-protein interactions have not been considered before in disease gene prioritization. Label propagation has been successfully applied to prioritize disease causing genes in previous network-based methods. These network-based methods use basic label propagation, i.e. random walk, on networks to prioritize disease genes in different ways. However, all these methods can not deal with the situation in which plenty false positive protein-protein interactions exist in the dataset, because the PPI network is used as a fixed input in previous methods. This important characteristic of data source may cause a large deviation in results. RESULTS: A novel network-based framework IDLP is proposed to prioritize candidate disease genes. IDLP effectively propagates labels throughout the PPI network and the phenotype similarity network. It avoids the method falling when few disease genes are known. Meanwhile, IDLP models the bias caused by false positive protein interactions and other potential factors by treating the PPI network matrix and the phenotype similarity matrix as the matrices to be learnt. By amending the noises in training matrices, it improves the performance results significantly. We conduct extensive experiments over OMIM datasets, and IDLP has demonstrated its effectiveness compared with eight state-of-the-art approaches. The robustness of IDLP is also validated by doing experiments with disturbed PPI network. Furthermore, We search the literatures to verify the predicted new genes got by IDLP are associated with the given diseases, the high prediction accuracy shows IDLP can be a powerful tool to help biologists discover new disease genes. CONCLUSIONS: IDLP model is an effective method for disease gene prioritization, particularly for querying phenotypes without known associated genes, which would be greatly helpful for identifying disease genes for less studied phenotypes. AVAILABILITY: https://github.com/nkiip/IDLP.

Human mobility analysis is one of the most important research problems in the field of urban computing. Existing research mainly focuses on the intra-city ground travel behavior modeling, while the inter-city air travel behavior modeling has been largely ignored. Actually, the inter-city travel analysis can be of equivalent importance and complementary to the intra-city travel analysis. Understanding massive passenger-air-travel behavior delivers intelligence for airlines’ precision marketing and related socioeconomic activities, such as airport planning, emergency management, local transportation planning, and tourism-related businesses. Moreover, it provides opportunities to study the characteristics of cities and the mutual relationships between them. However, modeling and predicting air traveler behavior is challenging due to the complex factors of the market situation and individual characteristics of customers (e.g., airlines’ market share, customer membership, and travelers’ intrinsic interests on destinations). To this end, in this article, we present a systematic study on the personalized air travel prediction problem, namely where a customer will fly to and which airline carrier to fly with, by leveraging real-world anonymized Passenger Name Record (PNR) data. Specifically, we first propose a relational travel topic model, which combines the merits of latent factor model with a neighborhood-based method, to uncover the personal travel preferences of aviation customers and the latent travel topics of air routes and airline carriers simultaneously. Then we present a multi-factor travel prediction framework, which fuses complex factors of the market situation and individual characteristics of customers, to predict airline customers’ personalized travel demands. Experimental results on two real-world PNR datasets demonstrate the effectiveness of our approach on both travel topic discovery and customer travel prediction.

We consider incorporating topic information into a sequence-to-sequence framework to generate informative and interesting responses for chatbots. To this end, we propose a topic aware sequence-to-sequence (TA-Seq2Seq) model. The model utilizes topics to simulate prior human knowledge that guides them to form informative and interesting responses in conversation, and leverages topic information in generation by a joint attention mechanism and a biased generation probability. The joint attention mechanism summarizes the hidden vectors of an input message as context vectors by message attention and synthesizes topic vectors by topic attention from the topic words of the message obtained from a pre-trained LDA model, with these vectors jointly affecting the generation of words in decoding. To increase the possibility of topic words appearing in responses, the model modifies the generation probability of topic words by adding an extra probability item to bias the overall distribution. Empirical studies on both automatic evaluation metrics and human annotations show that TA-Seq2Seq can generate more informative and interesting responses, significantly outperforming state-of-the-art response generation models.

Gene-phenotype association prediction can be applied to reveal the inherited basis of human diseases and help drug development. Gene-phenotype associations are related to complex biological process and influenced by various factors, such as relationship between phenotypes and that among genes. While due to sparseness of curated gene-phenotype associations, existing approaches are limited to prediction accuracy. In this paper, we propose a novel method by exploiting weighted graph constraint learned from hierarchical structures of phenotype data and group prior information among genes by inheriting advantages of Non-negative Matrix Factorization (NMF), called Weighted Graph Constraint and Group Centric Non-negative Matrix Factorization (GC2NMF). Specifically, firstly we introduce the depth of parent-child relationships between two adjacent phenotypes in hierarchal phenotypic data as weighted graph constraint for a better phenotype understanding. Secondly, we utilize intra-group correlation among genes in a gene group as group constraint for gene understanding. Such information provides us an intuitive priori that genes in a group probably result in similar phenotypes. The model allows not only to achieve a high prediction performance but also jointly to learn interpretable representation of genes and phenotypes to handle future biological analysis. Experimental results on biological gene-phenotype association datasets of mouse and human demonstrate that GC2NMF can obtain superior prediction accuracy and good understandability for biological explanation over other state-of-the-art methods.

Sub-event discovery is an effective method for social event analysis in Twitter. It can discover sub-events from large amount of noisy event-related information in Twitter and semantically represent them. The task is challenging because tweets are short, informal and noisy. To solve this problem, we consider leveraging event-related hashtags that contain many locations, dates and concise sub-event related descriptions to enhance sub-event discovery. To this end, we propose a hashtag-based mutually generative Latent Dirichlet Allocation model(MGe-LDA). In MGe-LDA, hashtags and topics of a tweet are mutually generated by each other. The mutually generative process models the relationship between hashtags and topics of tweets, and highlights the role of hashtags as a semantic representation of the corresponding tweets. Experimental results show that MGe-LDA can significantly outperform state-of-the-art methods for sub-event discovery.

In this paper, we introduce a new topic model to understand the chaotic microblogging environment by using hashtag graphs. Inferring topics on Twitter becomes a vital but challenging task in many important applications. The shortness and informality of tweets leads to extreme sparse vector representations with a large vocabulary. This makes the conventional topic models (e.g., latent Dirichlet allocation [1] and latent semantic analysis [2]) fail to learn high quality topic structures. Tweets are always showing up with rich user-generated hashtags. The hashtags make tweets semi-structured inside and semantically related to each other. Since hashtags are utilized as keywords in tweets to mark messages or to form conversations, they provide an additional path to connect semantically related words. In this paper, treating tweets as semi-structured texts, we propose a novel topic model, denoted as Hashtag Graphbased Topic Model (HGTM) to discover topics of tweets. By utilizing hashtag relation information in hashtag graphs, HGTM is able to discover word semantic relations even if words are not co-occurred within a specific tweet. With this method, HGTM successfully alleviates the sparsity problem. Our investigation illustrates that the user-contributed hashtags could serve as weakly-supervised information for topic modeling, and the relation between hashtags could reveal latent semantic relation between words. We evaluate the effectiveness of HGTM on tweet (hashtag) clustering and hashtag classification problems. Experiments on two real-world tweet data sets show that HGTM has strong capability to handle sparseness and noise problem in tweets. Furthermore, HGTM can discover more distinct and coherent topics than the state-of-the-art baselines.

Ranking SVM, which formalizes the problem of learning a ranking model as that of learning a binary SVM on preference pairs of documents, is a state-of-the-art ranking model in information retrieval. The dual form solution of Ranking SVM model can be written as a linear combination of the preference pairs, i.e., w = ∑(i,j) αij (xi - xj), where αij denotes the Lagrange parameters associated with each pair (i,j). It is obvious that there exist significant interactions over the document pairs because two preference pairs could share a same document as their items. Thus it is natural to ask if there also exist interactions over the model parameters αij, which we may leverage to propose better ranking model. This paper aims to answer the question. Firstly, we found that there exists a low-rank structure over the Ranking SVM model parameters αij, which indicates that the interactions do exist. Then, based on the discovery, we made a modification on the original Ranking SVM model by explicitly applying a low-rank constraint to the parameters. Specifically, each parameter αij is decomposed as a product of two low-dimensional vectors, i.e., αij = vi, vj, where vectors vi and vj correspond to document i and j, respectively. The learning process, thus, becomes to optimize the modified dual form objective function with respect to the low-dimensional vectors. Experimental results on three LETOR datasets show that our method, referred to as Factorized Ranking SVM, can outperform state-of-the-art baselines including the conventional Ranking SVM.

Mining topics in Twitter is increasingly attracting more attention. However, the shortness and informality of tweets leads to extreme sparse vector representation with a large vocabulary, which makes the conventional topic models (e.g., Latent Dirichlet Allocation) often fail to achieve high quality underlying topics. Luckily, tweets always show up with rich user-generated hash tags as keywords. In this paper, we propose a novel topic model to handle such semi-structured tweets, denoted as Hash tag Graph based Topic Model (HGTM). By utilizing relation information between hash tags in our hash tag graph, HGTM establishes word semantic relations, even if they haven't co-occurred within a specific tweet. In addition, we enhance the dependencies of both multiple words and hash tags via latent variables (topics) modeled by HGTM. We illustrate that the user-contributed hash tags could serve as weakly-supervised information for topic modeling, and hash tag relation could reveal the semantic relation between tweets. Experiments on a real-world twitter data set show that our model provides an effective solution to discover more distinct and coherent topics than the state-of-the-art baselines and has a strong ability to control sparseness and noise in tweets.

In this paper, we introduce “task trail” to understand user search behaviors. We define a task to be an atomic user information need, whereas a task trail represents all user activities within that particular task, such as query reformulations, URL clicks. Previously, web search logs have been studied mainly at session or query level where users may submit several queries within one task and handle several tasks within one session. Although previous studies have addressed the problem of task identification, little is known about the advantage of using task over session or query for search applications. In this paper, we conduct extensive analyses and comparisons to evaluate the effectiveness of task trails in several search applications: determining user satisfaction, predicting user search interests, and suggesting related queries. Experiments on large scale data sets of a commercial search engine show that: (1) Task trail performs better than session and query trails in determining user satisfaction; (2) Task trail increases webpage utilities of end users comparing to session and query trails; (3) Task trails are comparable to query trails but more sensitive than session trails in measuring different ranking functions; (4) Query terms from the same task are more topically consistent to each other than query terms from different tasks; (5) Query suggestion based on task trail is a good complement of query suggestions based on session trail and click-through bipartite. The findings in this paper verify the need of extracting task trails from web search logs and enhance applications in search and recommendation systems.

Capturing the context of a user's query from the previous queries and clicks in the same session leads to a better understanding of the user's information need. A context-aware approach to document reranking, URL recommendation, and query suggestion may substantially improve users' search experience. In this article, we propose a general approach to context-aware search by learning avariable length hidden Markov model(vlHMM) from search sessions extracted from log data. While the mathematical model is powerful, the huge amounts of log data present great challenges. We develop several distributed learning techniques to learn a very large vlHMM under themap-reduceframework. Moreover, we construct feature vectors for each state of the vlHMM model to handle users' novel queries not covered by the training data. We test our approach on a raw dataset consisting of 1.9 billion queries, 2.9 billion clicks, and 1.2 billion search sessions before filtering, and evaluate the effectiveness of the vlHMM learned from the real data on three search applications: document reranking, query suggestion, and URL recommendation. The experiment results validate the effectiveness of vlHMM in the applications of document reranking, URL recommendation, and query suggestion.

In this paper, we introduce "task trail" as a new concept to understand user search behaviors. We define task to be an atomic user information need. Web search logs have been studied mainly at session or query level where users may submit several queries within one task and handle several tasks within one session. Although previous studies have addressed the problem of task identification, little is known about the advantage of using task over session and query for search applications. In this paper, we conduct extensive analyses and comparisons to evaluate the effectiveness of task trails in three search applications: determining user satisfaction, predicting user search interests, and query suggestion. Experiments are conducted on large scale datasets from a commercial search engine. Experimental results show that: (1) Sessions and queries are not as precise as tasks in determining user satisfaction. (2) Task trails provide higher web page utilities to users than other sources. (3) Tasks represent atomic user information needs, and therefore can preserve topic similarity between query pairs. (4) Task-based query suggestion can provide complementary results to other models. The findings in this paper verify the need to extract task trails from web search logs and suggest potential applications in search and recommendation systems.

This paper is concerned with supervised rank aggregation, which aims to improve the ranking performance by combining the outputs from multiple rankers. However, there are two main shortcomings in previous rank aggregation approaches. Firstly, the learned weights for base rankers do not distinguish the differences among queries. This is suboptimal since queries vary significantly in terms of ranking. Besides, most current aggregation functions are unsupervised. A supervised aggregation function could further improve the ranking performance. In this paper, the significant difference existing among queries is taken into consideration, and a supervised rank aggregation approach is proposed. As a case study, we employ RankSVM model to aggregate the base rankers, referred to as Q.D.RSVM, and prove that Q.D.RSVM can set up query-dependent weights for different base rankers. Experimental results based on benchmark datasets show our approach outperforms conventional ranking approaches.

The ever increasing usage of acronyms in many kinds of documents, including web pages, is becoming an obstacle for average readers. This paper studies the task of finding expansions in documents for a given set of acronyms. We cast the expansion finding problem as a sequence labeling task and adapt Conditional Random Fields (CRF) to solve it. While adapting CRFs, we enhance the performance from two aspects. First, we introduce nonlinear hidden layers to learn better representations of the input data. Second, we design simple and effective features. We create a hand labeled evaluation data based on Wikipedia.org and web crawling. We evaluate the effectiveness of several algorithms in solving the expansion finding problem. The experimental results demonstrate that the new method achieves performs better than Support Vector Machine and standard Conditional Random Fields.

Recently, combining Conditional Random Fields (CRF) with Neural Network has shown the success of learning high-level features in sequence labeling tasks. However, such models are difficult to train because of the increase of the parameters to tune which needs enormous of labeled data to avoid over fitting. In this paper, we propose a transfer learning framework for the sequence labeling task of gesture recognition. Taking advantage of the frame correlation, we design an unsupervised sequence model as a pseudo auxiliary task to capture the underlying information from both the labeled and unlabeled data. The knowledge learnt by the auxiliary task can be transferred to the main task of CRF with a deep architecture by sharing the hidden layers, which is very helpful for learning meaningful representation and reducing the need of labeled data. We evaluate our model under 3 gesture recognition datasets. The experimental results of both supervised learning and semi-supervised learning show that the proposed model improves the performance of the CRF with Neural Network and other baseline models.

A cooperation approach with consideration of communication limit is proposed for multi-robot area exploration, in which all the robots select local destinations satisfying the constraints on communication range and reach their destinations at the same time to communicate and fuse their map information. Firstly, the robots compute the frontier between the explored region and the unexplored one. The robots choose the optimal frontier points, which maximize information gain, minimize navigation cost and satisfy communication limit as their local destinations. Then the problem of global exploration in unknown environment is converted into that of multi-stage trajectory planning in local known environment. Collision-free, synchronous and separate trajectories are planned for all the robots to realize the limited communication at their destinations. In such a way, efficient and distributed exploration can be achieved. Simulation results are presented to show the effectiveness of our method.

This paper investigates the on-line autonomous distribution for landmarks and the active exploration in environment without or lack of landmarks/features, such as disaster conditions and polar region. In such situation, the robot enters the environment carrying some landmarks and distributes them according to the rules given in this paper. The utility of the landmark distribution is analyzed. Then, based on the extended Kalman filter (EKF), the active exploration is converted into a problem of multi-objective optimization, in which the objective function includes three aspects, i.e. the accuracy of localization and mapping, the predictive area of the unknown environment that will be explored in next step and the information gain provided by the distributed landmarks respectively. The robot chooses the control input that optimizes the objective function such that accurate localization, high-quality mapping and complete exploration will be realized. And then, the supplementation and the redundancy elimination for landmarks are implemented. At last, a set of simulations is presented to show the effectiveness of our approach.

This paper points out that many machine learning problems in IR should be and can be formalized in a novel way, referred to as 'group-based learning'. In group-based learning, it is assumed that training data as well as testing data consist of groups. The classifier is created and utilized across groups. Furthermore, evaluation in testing and also in training are conducted at group level, with the use of evaluation measures defined on a group. This paper addresses the problem and presents a Boosting algorithm to perform the new learning task. The algorithm, referred to as AdaBoost.Group, is proved to be able to improve accuracies in terms of group-based measures during training.

This paper proposes a new ranking approach for information retrieval, where the diversity among queries are taken into consideration. In information retrieval, the users' queries often vary a lot from one to another, so that the documents retrieved from different queries are also distributed differently. Due to this diversity, it is not appropriate to assume all the documents to be ranked are generated i.i.d. (independently and identically distributed) according to a fixed but unknown probability distribution. However, most of the existing learning to rank approaches are proposed on the basis of the conventional i.i.d. assumption. In this paper, the conventional i.i.d. assumption is relaxed to fit the real situations of information retrieval better, and then a new ranking approach, referred to as 'query dependent ranking', is proposed. In our approach, the ranking models for different queries have generality while each of them has its own speciality. The experimental results on both synthetic and real-world datasets show the advantage of our approach to conventional ranking approaches.

DJ DreamFactory is a web-based integrated platform for interactive broadcasting over the Internet. In DJ DreamFactory, users are able to set up Internet-based broadcasting stations with minimal effort. Audience not only can receive broadcasting programs from multiple stations through this unified platform, but can also "talk" and "write" to the broadcasters as well as to other audience via real-time text and/or audio interactions.

In this paper, we present a web-based aggregated platform, DJ DreamFactory, which enables average users to effortlessly participate in and contribute to interactive media broadcasting over the Internet. The platform overcomes several shortcomings of existing Internet-based broadcasting systems, such as inconvenience in channel surfing and content browsing due to the scattering and isolating of broadcasting stations, difficulties in setting up a broadcasting station, lack of communications between broadcasters and audience, and little support for personalized experience. The proposed platform facilitates users' media access by seamlessly aggregating sporadic broadcasting stations run by individual hosts, and enables a virtual community where grassroots users can contribute to media broadcasting, sharing, organizing and annotating through social networking. In addition, it supports real-time multimodal interaction between audience and hosts, provides customized services for both broadcasters and audience, supports personalized media experiences by mining and managing audience's preferences, and facilitates the organization of unstructured media data collections as well as collective human intelligence on the Web.

This paper1 is concerned with the problem of multimodal fusion in video search. First, we employ an object-sensitive approach to query analysis to improve the baseline result of text-based video search. Then, we propose a PageRank-like graph-based approach to text-based search result re-ranking. To better exploit the underlying relationship between video shots, the proposed re-ranking scheme simultaneously leverages textual relevancy, semantic concept relevancy, and low-level-feature-based visual similarity. In this PageRank-like scheme, we construct a set of graphs with the video shots as vertexes, and the conceptual and visual similarity between video shots as "hyperlinks". A modified topic-sensitive PageRank algorithm is then applied on these graphs to propagate the relevance scores through all related video shots. Experimental results verify the effectiveness of the graph-based propagation approach combined with the object-sensitive query analysis approach, which brings significant improvement to the baseline of text-based video search. Our experimental analysis also indicates that the proposed re-ranking method is highly generic and independent of different query classes, training data, and human interference.

This paper addresses the problems of the forward and backward path following control for tractor-trailer mobile robots (TTMR) with connection structures of on-axle hitching and off-axle hitching. First, the kinematics is described and the motion characteristics are analyzed. Then, by Lyapunov method we design a global path following controller for single-body mobile robot and extend it to the forward path following control of the TTMR. Furthermore, by the kinematics transformation and the backstepping technique respectively, we propose approaches to the backward path following control for TTMR with two kinds of connection structures based on the above controller. Finally, a set of simulations is presented to show the validity of our approach

Path planned for tractor-trailer mobile robot (TTMR) does not seem to be easily adapted to the case when the number of the trailers increases, thus the repetitive path planning has to be introduced. For this problem, this paper investigates the connection relation of TTMR with variable structure, i.e. the number of the trailers is variable, and deals with the optimization design for it to avoid the repetitive path planning when the number of the trailers increases. After analyzing the motion trajectories of TTMR, we establish the quantitative relationship for the motion characteristics between the transient course and the steady state. Based on which, a new design method to optimize the connection relation for variable-structure TTMR is proposed. In our scheme, the length of each connecting rod between two adjacent bodies is adjusted properly such that the transient deviations of the trailers from the path followed by the tractor will be less than the steady ones. In such a way, the path planned for TTMR can also be appropriate for the case when the number of the trailers varies. The simulations and experiments show the validity of our method

The paper is concerned with applying learning to rank to document retrieval. Ranking SVM is a typical method of learning to rank. We point out that there are two factors one must consider when applying Ranking SVM, in general a "learning to rank" method, to document retrieval. First, correctly ranking documents on the top of the result list is crucial for an Information Retrieval system. One must conduct training in a way that such ranked results are accurate. Second, the number of relevant documents can vary from query to query. One must avoid training a model biased toward queries with a large number of relevant documents. Previously, when existing methods that include Ranking SVM were applied to document retrieval, none of the two factors was taken into consideration. We show it is possible to make modifications in conventional Ranking SVM, so it can be better used for document retrieval. Specifically, we modify the "Hinge Loss" function in Ranking SVM to deal with the problems described above. We employ two methods to conduct optimization on the loss function: gradient descent and quadratic programming. Experimental results show that our method, referred to as Ranking SVM for IR, can outperform the conventional Ranking SVM and other existing methods for document retrieval on two datasets.

Coordination control for multiple robot manipulators is investigated. The controlled variables, i.e., the external and internal arm forces, are first defined, based on the decomposition of the robot arm force, and then a hybrid control scheme at the robot end-effector level is proposed. The control scheme consists of three loops: (i) an external force loop, which enables the robot manipulators to induce the desired object dynamics and contact force between the object and the environment, (ii) an internal force loop, which makes the manipulators impose the desired internal force on the object, and (iii) a robot motion loop which controls the robot arms to track the desired trajectories. A criterion related to the load capability of the robot arm is presented for distributing a load among the manipulators. An integrated control system for two industrial robots is developed.