The Study Of Association Rule Mining Based On Concept Lattice

Nowadays, the society has entered the information era. The computer and network information technology have been developed so rapidly that data and information are increasing dramatically (information explosion) in all fields, meanwhile data and information systems become more uncertain due to human's participation. How to effectively achieve the data analysis and processing and quickly get implicit knowledge has long been an important direction of Artificial intelligence. In this background, Knowledge Discovery in Databases (KDD) and Data Mining (DM) provide a new intelligent way of understanding data.The association rules which are an important pattern in data mining have widely investigated. Concept Lattice with complete structure and good theory has been one of the main tools in DM. Therefore it is significant to study association rules by using of concept Lattice.Based on Rough Set Theory and Concept Lattice, a new method is proposed to search for association rules. The formal context is first reduced via Rough Set Theory. And then some of the attributes are thrown off for the given threshold. The complexity of constructing lattice and searching for the desire concept is deceased. The supports and confidences for the obtained association rules are computed with the aid of concept lattice. The experiment shows the validity and detail steps of this method.