Multistrategy Data Exploration Using the INLEN System: Recent Advances
Ryszard S. Michalski, Kenneth A. Kaufman · 1997
Recent advances in the development of the INLEN system for multistrategy data exploration are briefly reviewed. These advances include the development of a meta-level language for data mining and knowledge discovery, called knowledge generation language (KGL), and the employment of a new type of attributes, called structured attributes. These features are illustrated by an example concerned with determining economic and demographic patterns in a database containing facts about the countries of the world. The results demonstrate a high utility of INLEN for data mining and knowledge discovery. Introduction The availability of very large volumes of data in the electronic form has created a problem of deriving from them useful, task-oriented knowledge. Traditional data analysis techniques, which include statistical and numerical methods, are oriented primarily toward the extraction of quantitative data characteristics, and as such have inherent limitations. For example, statistical techn...