On the Extraction and Matching Between Structural Expertise and Fuzzy-Based Description of Objects

The representation of object features is an important task in the researches such as pattern recognition, information retrieval, interactive query, etc. This paper addresses how to integrate computational linguistics and fuzzy set techniques to automatically establish the knowledge base for semi-structural domain expertise. The proposed task includes document pre-processing, lexical vector encoding, fuzzy vector shrinking and fuzzy membership encoding. The content of a wild bird illustrated book was used as the training corpus and the domain expertise was established. Queries from another book, an expert and a naive were used as testing corpus. The preliminary results showed that the proposed approach is suitable for representing domain expertise. The top-N scores an average of 79 query for a specific object.