ON DATA MINING AND DATA WAREHOUSING FOR MULTIMEDIA INFORMATION RETRIEVAL

摘要

This paper describes a new approach to fast multimedia information retrieval with data mining and data warehousing techniques. To tackle the key issues such as multimedia data indexing, similarity measures, search methods and query processing in retrieval for large multimedia data archives, we extend the concepts of conventional data warehouse and multimedia database to multimedia data warehouse for effective data representation and storage. In addition, we propose a fuzzy neural network to provide automatic and autonomous classification for the retrieval outputs by integrating fuzzy logic technology and the Back Propagation Feed Forward (BPFF) neural network. A series of case studies are reported to demonstrate the feasibility of the proposed method.