A framework for quick-and-pinpoint data mining and its application to heterogeneous genome databases

摘要

This paper proposes a new framework for knowledge discovery form large and heterogeneous genome databases. By regarding a set of entries as user's interest, the system drastically cuts off irrelevant items and transactions to allow "quick-and-pinpoint" data mining. Two examples of practical applications showed that the framework is effective for knowledge discovery from genome databases via the WWW.