Ontology Mapping Between Heterogeneous Product Taxonomies in an Electronic Commerce Environment

摘要

Ontology merging and mapping is one of the most important research areas in the Semantic Web. Electronic commerce requires ontology mapping on product comparison over the different product classification taxonomies of shopping malls. Since a strict mapping strategy may lead a customer's configuration to search failure, a mapping algorithm for product matching is suggested that can provide more products by increasing recall with reasonable loss of precision. The algorithm was compared with PROMPT in six experimental sets. On average, the algorithm's recall was better than PROMPT's by 46.5 percent, while its precision was worse than PROMPT's by 13.6 percent. The evaluation confirmed that the objective of significantly increased recall with reasonably decreased precision was successfully achieved.