An Algorithm for Ontology Matching in P2P E-commerce Environments Based on Semantic Affinity

摘要

Recommendation systems are applied to personalize and customize the E-Commerce Web environment. Similarities and difference among mainstream recommendation technology and system are analyzed. The fundamental challenges for recommendation system based on collaborative filtering (CF) technology are the scalability and accuracy in real-time. By giving to definitions of SPCCF (Semantic Peer-to-Peer Community for Collaborative Filtering) and peer-ontology, compared to the H-match Algorithm for ontology matching in informational retrieval, linguistic and semantic affinity weight were discussed. By hybrid considering goods ontology concept recommended in SPCCF and Peer Structure (including related concepts and property) Semantic Affinity, a hybrid algorithm for ontology matching was presented. At last an instance proved this algorithm's accuracy.