Dantong Ouyang

Integrity constraints in OWL ontologies based on grounded circumscription

The extensions for logic-based knowledge bases with integrity constraints are rather popular. We put forward an alternative criteria for analysis of integrity constraints in Web ontology language (OWL) ontology under the closed world assumption. According to this criteria, grounded circumscription is applied to define integrity constraints in OWL ontology and the satisfaction of the integrity constraints by minimizing extensions of the predicates in integrity constraints. According to the semantics of integrity constraints, we provide a modified tableau algorithm which is sound and complete for deciding the consistency of an extended ontology. Finally, the integrity constraint validation is converted into the corresponding consistency of the extended ontology. Comparing our approach with existing integrity constraint validation approaches, we show that the results of our approach are more in accordance with user requirements than other approaches in certain cases.

Solving SAT Problem with Boolean Algebra

The satisfiability (SAT) problem is an important problem of automated reasoning. In the past decades, many methods of SAT are proposed, such as method based on resolution, method based on tableau and method based on extension rule. Based on the study of extension rule, we find that the SAT problem can be solved with hitting set algorithms. If we can find a hitting set of a clause set, and if there is not any pair of complementary literals in the hitting set, the clause set is satisfiable. Algorithm BHS based on Boolean algebra proposed by Jiang is an efficient hitting set algorithm. SSBF proposed in this paper is a revised algorithm of BHS that can be used to judge the satisfiability of a clause set. The test result shows that SSBF is an efficient algorithm.

Integrity Constraint Validation in DL-Lite _R Based Ontology Using Rewriting

With the rapid development of information technology, semantic web data present features of massiveness and complexity. As the data-centric science, social computing have great influence in collecting and analyzing semantic data. In our contribution, we propose an integrity constraint validation for DL-Lite \(_R\) based ontology in view of data correctness issue in the progress of social computing applications. Firstly, at the basis of translations from integrity constraint axioms into a set of conjunctive queries, integrity constraint validation is converted into the conjunctive query answering over knowledge bases. Moreover, rewriting rules are used for reformulating the integrity constraint axioms using standard axioms. On this account, the integrity constraint validation can be reduced to the query evaluation over the ABox, and use query mechanisms in database management systems to optimize integrity constraint validation. Finally, the experimental result shows that the rewriting-based method greatly improves the efficiency of integrity constraint validation and is more appropriate to scalable data in the semantic web.

Distant Supervision for Relation Extraction with Type Constraint

Distant supervision has been proposed for extracting relation instances from unstructured corpus. However, it is based on the assumption that each sentence mentioning two related entities is an expression of the given relation, which is not always the case and in turn can lead to noisy patterns and thus hurt precision. To improve the accuracy, this paper proposes an approach by adding entitys information as extra type constraint into the distance supervision based framework to train a relation extractor. To achieve this, a way to align knowledge base entities to their text mentions, and extract each entitys information (features) from those text mentions are provided in first. Secondly, two ways of adding entitys information to the problem of relation extraction are explored: (1) a joint algorithm, which models entitys features and relations features jointly, (2) a type-constraint algorithm, which uses such features to constrain the types of relation arguments. The experimental results demonstrate that our approach can significantly improve the accuracy on the interesting datasets.

Solving SAT problem with a revised hitting set algorithm

The satisfiability(SAT) problem is a core problem of artificial intelligence. Research findings in SAT are widely used in many areas. The main methods solving SAT problem are resolution principle, tableau calculus and extension rule. Besides methods mentioned above, we find that the SAT problem can be solved with hitting set algorithms. If a set of clause is satisfiable, there must be a hitting set of the clause set which containing no complementary pairs of literals. Algorithm NEWHS-tree is an efficient hitting set algorithm proposed by Ouyang. RNHST proposed in this paper is a revised algorithm in respect of NEWHS-tree. It judges the satisfiability of a clause set by confirming the existence of the sets hitting set without complementary pairs of literals. The test result shows that RNHST is an efficient algorithm.

An algorithm based on resolution for the satisfiability problem

The satisfiability problem is the core problem in artificial intellgence. The algorithm directional resolution(DR) is a well known method based on resolution for satisfiability problem. But the number of clauses has great impact on the efficiency of the method. In this paper, an algorithm SRDR is proposed to solve the problem. SRDR is based on algorithm DR and splitting rule. By using splitting rule, the number of clauses can be reduced obviously. Furthermore, the strategy MO is designed for SRDR. With the strategy, we can get a better order of variables and the efficiency of SRDR is improved. The experimental data shows that SRDR is more efficient that DR.

Persistent Storage of Ontology-Based Data with Integrity Constraints

Recently, a lot of approaches and systems were proposed to store the ontology-based data to the relational database. In this paper, we present a novel approach to store the ontology-based data with integrity constraints. Integrity constraint (IC) is a formula that checks whether all the necessary information has been explicitly provided. It can be added into ontology to guarantee the integrity of the ontology. A set of constraint axioms called IC-based axioms are proposed to represent the integrity constraints in the ontology and further store the ontology-based data in the relational database, which satisfy the integrity constraints. Through comparing with other existing approaches used in Semantic Web applications, the experiment shows that our approach is advanced to the traditional approaches.