In this paper, we propose a hybrid model that combines neural network and hidden Markov model for time-aware recommender systems. We use higher-order hidden Markov model to capture the temporal information of users and items in collaborative filtering systems. Because the computation of the transition matrix of higher-order hidden Markov model is hard, we compute the transition matrix by deep neural networks. We implement the algorithms of the hybrid model for offline batch-learning and online updating respectively. Experiments on real datasets demonstrate that the hybrid model has improvement performances over the existing recommender systems.

In Reiter’s default logic, it is possible that no useful information can be brought from inconsistent knowledge or no extension of incoherent default theories exists. In this paper, based on Belnap’s four-valued logic, we propose a new variant of default logic called the restricted four-valued default logic to tolerate inconsistency and incoherency of knowledge in default reasoning. Our proposal can maintain both the expressive power of full default logic and the ability of default reasoning. Moreover, we present a transformation-based approach to compute the restricted four-valued extensions.

Recommender system has become an effective tool for information filtering, which usually provides the most useful items to users by a top-k ranking list. Traditional recommendation techniques such as Nearest Neighbors (NN) and Matrix Factorization (MF) have been widely used in real recommender systems. However, neither approaches can well accomplish recommendation task since that: (1) most NN methods leverage the neighbor's behaviors for prediction, which may suffer the severe data sparsity problem; (2) MF methods are less sensitive to sparsity, but neighbors' influences on latent factors are not fully explored, since the latent factors are often used independently. To overcome the above problems, we propose a new framework for recommender systems, called collaborative factorization. It expresses the user as the combination of his own factors and those of the neighbors', called collaborative latent factors, and a ranking loss is then utilized for optimization. The advantage of our approach is that it can both enjoy the merits of NN and MF methods. In this paper, we take the logistic loss in RankNet and the likelihood loss in ListMLE as examples, and the corresponding collaborative factorization methods are called CoF-Net and CoF-MLE. Our experimental results on three benchmark datasets show that they are more effective than several state-of-the-art recommendation methods.

Role based access control (RBAC) has emerged as a leading access control model to other traditional access control models. However, the traditional RBAC models can not capture fine-grained authorization with mono-type inheritance. In this paper, we discuss the hybrid inheritance based on our extended RBAC model, which is very desirable for capturing the fine-grained access control permissions. When the hybrid inheritances coexist in a role hierarchy, inferring such indirect relations between a pair of roles can became very complex. In particular, we study how the new inheritance relations between roles that are indirectly related can be easily derived through the inference rules, which provides a basis for formally analyzing the hybrid inheritances.

Uniting ontologies and rules has become a central topic in the Semantic Web. Bridging the discrepancy between these two knowledge representations, this paper introduces DatalogDL as a family of hybrid languages, where Datalog rules are parameterized by various DL (description logic) languages ranging from to . Making DatalogDL a decidable system with complexity of EXPTIME, we propose independent properties in the DL body as the restriction to hybrid rules, and weaken the safeness condition to balance the trade‐off between expressivity and reasoning power. Building on existing well‐developed techniques, we present a principled approach to enrich (RuleML) rules with information from (OWL) ontologies, and develop a prototype system combining a rule engine (OO jDREW) with a DL reasoner (RACER).

In this paper we describe the paraconsistent circumscription by the application of predicate circumscription in a paraconsistent logic, the logic of paradox LP. In addition to circumscribing the predicates, we also circumscribe the inconsistency. The paraconsistent circumscription can be well characterized by the minimal semantics which is both nonmonotonic and paraconsistent. It brings us advantages in two respects: nonmonotonic logic would be nontrivial while there was a contradiction, and paraconsistent logic would be equivalent to classical logic while there was no effect of a contradiction.