Concept similarity analysis in Ontology´s automatic extraction

Ontology´s automatic extraction is a core problem of information integration in electronic government affair. In the process of ontology´s automatic extraction, FCA method is used in analyzing relationships between concepts automatically. But this method´s ability is insufficient in the analysis of the synonym relationship. This paper optimizes the FCA method and brings forward a new algorithm - SFCA. SFCA sets the weight for the attribute based on the importance of it. It computes the similarity degree using the weights and judges whether the concepts are synonymous. Through the analysis of the experiment´s result, the algorithm is validated to be effective. And its correctness proof is proved.