An Incremental Approach for Attribute Reduction in Concept Lattice

摘要

One of the key problems of knowledge discovery is knowledge reduction. Concept lattice is an effective tool for data analysis and knowledge processing. The existing works on attribute reduction in concept lattice have mainly been focused on static database. This paper presents an incremental approach to identify reductions from dynamic database. The properties of attributes are discussed within the framework of equivalence classes and the determinant theorem of attribute reduction is presented. Based on the theorem, the reductions can be easily derived. The experimental results validate the effectiveness of the approach.