A Novel Hybrid Approach for Incomplete Knowledge System Mining Based on Approximate Rough Entropy Lattice

Provided by: Binary Information Press
Topic: Data Management
Format: PDF
Based on some special advantages of both rough set analysis in knowledge reduction and formal concept analysis in formal concept representation, a novel hybrid approach for incomplete knowledge system mining based on approximate rough entropy concept is presented in this paper. First of all, the incomplete information system is transformed into the complete knowledge one by the attribute reduction of condition entropy. Secondly, the approximate condition entropy lattice mining model is constructed and all approximate knowledge rules are extracted from super-concept sub-concept relations implied in the formal context.

Find By Topic