Recommender system has become an indispensable component in many e-commerce sites. One major challenge that largely remains open is the cold-start problem, which can be viewed as an ice barrier that keeps the cold-start users/items from the warm ones. In this paper, we propose a novel rating comparison strategy (RaPare) to break this ice barrier. The center-piece of  our RaPare is to provide a fine-grained calibration on the latent profiles of cold-start users/items by exploring the differences between cold-start and warm users/items. We instantiate our RaPare strategy on the prevalent method in recommender system, i.e., the matrix factorization based collaborative filtering. Experimental evaluations on two real data sets validate the superiority of our approach over the existing methods in cold-start scenarios.

Large sparse datasets are common in many real-world applications. Linear SVM has been shown to be very efficient for classifying such datasets. However, it is still unknown how data sparseness would affect its convergence behavior. To study this problem in a systematic manner, we propose a novel approach to generate large and sparse data from real-world datasets, using statistical inference and the data sampling process in the PAC framework. We first study the convergence behavior of linear SVM experimentally, and make several observations, useful for real-world applications. We then offer theoretical proofs for our observations by studying the Bayes risk and PAC bound. Our experiment and theoretic results are valuable for learning large sparse datasets with linear SVM.

Microarray experiments are capable of measuring the expression level of thousands of genes simultaneously. Dealing with this enormous amount of information requires complex computation. Support Vector Machines (SVM) have been widely used with great efficiency to solve classification problems that have high dimension. In this sense, it is plausible to develop new feature selection strategies for microarray data that are associated with this type of classifier. Therefore, we propose, in this paper, a new method for feature selection based on an ordered search process to explore the space of possible subsets. The algorithm, called Admissible Ordered Search (AOS), uses as evaluation function the margin values estimated for each hypothesis by a SVM classifier. An important theoretical contribution of this paper is the development of the projected margin concept. This value is computed as the margin vector projection on a lower dimensional subspace and is used as an upper bound for the current value of the hypothesis in the search process. This enables great economy in runtime and consequently efficiency in the search process as a whole. The algorithm was tested using five different microarray data sets yielding superior results when compared to three representative feature selection methods.

Baader et al., 2013; Borgwardt et al., 2013; Özcep et al., 2013; Klarman and Meyer, 2014] and shown to preserve query Aiming at ontology-based data access over temporal, rewritability. Note, however, that the inability to define temporal in particular streaming data, we design a language of predicates such as Blizzard(x, t) in ontologies leaves the ontology-mediated queries by extending OWL 2 QL burden of encoding them within queries to the user, which and SPARQL with temporal operators, and investigate goes against the OBDA paradigm. Moreover, natural queries rewritability of these queries into two-sorted such as'check if a weather station has been serviced every 24 first-order logic with and PLUS over time.

The adoption of a generic contrariness notion in ASPIC+ substantially enhances its expressiveness with respect to other formalisms for structured argumentation. In particular, it opens the way to novel investigation directions, like the use of multivalued logics in the construction of arguments. This paper points out however that in the current version of ASPIC+ a serious technical difficulty related with generic contrariness is present. With the aim of preserving the same level of generality, the paper provides a solution based on a novel notion of closure of the contrariness relation at the level of sets of formulas and an abstract representation of conflicts between sets of arguments. The proposed solution is shown to satisfy the same rationality postulates as ASPIC+ and represents a starting point for further technical and conceptual developments in structured argumentation.

In this paper we focus on the greedy matrix completion problem. A simple atom selection strategy is proposed building upon an alternating minimization procedure. Based on this per-iteration strategy, we devise a greedy algorithm OAMC and establish an upper bound of the approximation error. To evaluate different weight refinement methods, several variants of OAMC are designed. We prove that OAMC and three of its variants have the property of linear convergence. Experiments of Recommendation and Image Recovery are conducted to make empirical evaluation. Results are promising. We report that our algorithm takes only 700 seconds to process Yahoo Music dataset in PC, yet achieves a root mean square error 24.5 on test set.

As part of the ongoing effort of designing secure DCOP algorithms, we propose P-Max-Sum, the first private algorithm that is based on Max-Sum. The proposed algorithm has multiple agents preforming the role of each node in the factor graph, on which the Max-Sum algorithm operates. P-Max-Sum preserves three types of privacy: topology privacy, constraint privacy, and assignment/decision privacy.By allowing a single call to a trusted coordinator, P-Max-Sum also preserves agent privacy. The two main cryptographic means that enable this privacy preservation are secret sharing and homomorphic encryption. Our experiments on structured and realistic problems show that the overhead of privacy preservation in terms of runtime is reasonable.

User interest inference from social networks is a fundamental problem to many applications. It usually exhibits dual-heterogeneities: a user's interests are complementarily and comprehensively reflected by multiple social networks; interests are inter-correlated in a nonuniform way rather than independent to each other. Although great success has been achieved by previous approaches, few of them consider these dual-heterogeneities simultaneously. In this work, we propose a structure-constrained multi-source multi-task learning scheme to co-regularize the source consistency and the tree-guided task relatedness. Meanwhile, it is able to jointly learn the task-sharing and task-specific features. Comprehensive experiments on a real-world dataset validated our scheme. In addition, we have released our dataset to facilitate the research communities.

Justification logic originated from the study of the logic of proofs.  However, in a more general setting, it may be regarded as a kind of explicit epistemic logic. In such logic, the reasons why a fact is believed are explicitly represented as justification terms.  Traditionally, the modeling of uncertain beliefs is crucially important for epistemic reasoning. While graded modal logics interpreted with possibility theory semantics have been successfully applied to the representation and reasoning of uncertain beliefs, they cannot keep track of the reasons why an agent believes a fact. The objective of this paper is to extend the graded modal logics with explicit justifications. We introduce a possibilistic justification logic,  present its syntax and semantics, and investigate its meta-properties, such as soundness, completeness, and realizability.

The semantic class (i.e., type) of an entity plays a vital role in many natural language processing tasks, such as question answering. However, most of existing type classification systems extensively rely on hand-crafted features. This paper introduces a hybrid neural model which classifies entity mentions to a wide-coverage set of 22 types derived from DBpedia. It consists of two parts. The mention model uses recurrent neural networks to recursively obtain the vector representation of an entity mention from the words it contains. The context model, on the other hand, employs multilayer perceptrons to obtain the hidden representation for contextual information of a mention. Representations obtained by the two parts are used together to predict the type distribution. Using automatically generated data, these two parts are jointly learned. Experimental studies illustrate that the proposed approach outperforms baseline methods. Moreover, when type information provided by our method is used in a question answering system, we observe a 14.7% relative improvement for the top-1 accuracy of answers.