Weighted Incremental Construction Of The Concept Lattice And Classification Rule Extraction

Weighted concept lattice is a concept lattice structure for the importance of attribute, by introducing weight into concept lattice's intension. Weighted concept lattice ensure the completeness of the lattice by introducing the concept of virtual node. In this paper, the incremental constructing algorithm and classification rule acquisition are studied based on frequent weighted concept lattice.The main research works can be summarized as follows:First, an incremental constructing algorithm (PWCL) of frequent weighted concept lattice is presented based on pruning. By using the rigorous monotone relation between father concept's intent and child concept's intent ,and setting the weight threshold by user, redundant comparative operations between the intents are eliminated by using pruning technology, frequent nodes and virtual nodes are generated,so that the efficiency of constructing lattice is improved and the completeness of lattice structure is ensured. In the end, experimental results validate that the correctness and effectiveness of the algorithm by taking the star spectrum data as the formal context..Second, a weighted classification rule extraction algorithm CRAAFWCL is presented. By using weighted concept lattice as a classification rule extraction tool, a classification rule extraction algorithm (CRAAFWCL) is presented by introducing concept of the weighted extant support. In the end, experimental results validate that the algorithm has higher classification efficiency by taking the star spectrum data as the formal context.