Research on Semantic Similarity Calculation of Linked Data Based on Multiple Factors

Semantic similarity calculation has been an important role in information retrieval of linked data, so calculation results will directly affect data mining results. To solve the problem of lower computation precision caused by the unity of factors on the research of semantic similarity calculation on linked data and the underutilization on semantic information of concept, this paper proposes a new semantic similarity calculation method based on multiple factors. This method combines the importance of attribute, types of attribute value with correlation. Firstly, it assigns the corresponding weight to the attribute, and uses the matching similarity algorithm of attributes according to the types of attribute value, and then similarity computation based on concept attributes is done. Secondly, it defines the path of correlation and determines the length of path, and then similarity computation based on correlation is done. Finally, through integrating the results of computation, the more accurate similarity can be get The experiment confirms that the proposed method fully utilizes semantic information of the concept and the calculation result can better reveal the similarity relation between concepts in linked data, compared with the calculation results based on the unity of factors.