Research On Semantic Similarity Of Ontology

In today’s era of big data, how accurate retrieved from the flood of information to meet the needs of information has become the primary task in the field of information retrieval. Semantic Web is semantic information retrieval technology and information retrieval research areas are linked closely together, it aims to provide a highly efficient and intelligent information retrieval system. Because of semantic heterogeneity problem, semantic information retrieval based on the traditional information retrieval methods using keyword matching and traditional semantic query expansion has been unable to meet user demand deep. Ontology proposed semantic information retrieval provides a new way. By ontology semantic similarity computing research to solve semantic heterogeneity. The semantic similarity calculation and Naive Bayes algorithm for semantic query expansion to achieve integrated semantic query expansion. This paper focuses on research and discussion in the following areas:First, the properties and the calculation method for the calculation of the existing lack of similarity based on semantic ontology-based information to improve the content based on a distance-based multi-factor ontology semantic similarity. Experiments show that improved calculation method to enhance the accuracy of the results.Second, the body multivariate semantic similarity based on the study, the use of principal component analysis to calculate the weight of each factor is given a comprehensive adaptive weighted algorithm that compared with traditional semantic similarity algorithm to calculate the accuracy of the algorithm with a big margin improvement. Comprehensive ontology semantic similarity adaptive weighted algorithm(ACWA) presented the results of the calculation accuracy of the reference standard value of Pearson coefficient average of 12.1%.Third, in order to solve the ontology-based semantic query expansion "Query drift" and proposed a comprehensive semantic query expansion method, taking into account the query word and the concept of the body does not match the problem, anaive Bayes calculate correlation, binding based on ontology semantic query expansion comprehensive semantic query expansion. Through experimental verification of integrated semantic query the proposed algorithm. Comparative experimental results show that the proposed method is superior to other query performance query expansion.