Building and Querying Large Modelbases

Abstract

Model building is one of the most important objectives of data mining and data analysis. As many data mining applications, such as personalization, bioinformatics and some large enterprise-wide business applications, become increasingly complex and require a very large number of models, it is becoming progressively more difficult for data analysts to built and to manage a large number of models in these applications on their own. Therefore, development of software tools helping data analysts in these tasks is becoming a pressing issue. This paper presents a model management system supporting various types of data mining models. It describes how to build and populate large heterogeneous modelbases. It also presents a query language for querying these modelbases and examines performance results for some of the queries.