Toward a Multi-Strategy and Cooperative Discovery System

We have been developing a methodology/system called GLS (Global Learning Scheme) for knowledge discovery in databases. The development of GLS has two main aspects. The first is to develop a m&i-strategy system. That is, many kinds of discovery/learning methods are cooperatively used in multiple learning phases for performing multi-aspect intelligent data analysis as well as multi-level conceptual abstraction and learning. As a multi-strategy system, GLS is implemented as a toolkit composed of several sub-systems and optional parts with a multi-level structure. We have finished main parts belonging to this aspect, and have undertaken another aspect, i.e., extending GLS into a multi-agent, distributed and cooperative discovery system. We try to increase versatility and autonomy of GLS by multi-strategy and distributed cooperation, This paper briefly discusses these two aspects of GLS.