Research On Effect-Based Roughness Measure Theory And Methods

With the rapid development of computer technology,the arrival of information age makes the data from all walks of life especially Internet increase dramatically.Rough set theory,as a kind of mathematical theory to process inconsistent,uncertainty or incomplete knowledge,not only attracts more and more attention in the academia,but also has been widely used in industry.Rough set theory,who is an important kind of intelligent effective processing tool for data mining,machine learning,knowledge discovery and decision support and many other fields,has become a hot field in the theory and application study of information science.For the roughness measure problems of rough sets,this paper firstly discusses the characteristics and shortcomings of the existing Pawlak roughness measure methods.Following this,a new effect-based roughness measure model is established using a combination of a basic measure factor,the lower(or upper)accuracy of rough sets,entitled the effective rough degree(abbreviated as ERD).Next,the characteristics of ERD are analyzed in combination with different synthesis functions,and several further necessary and sufficient conditions are given.At the same time,we propose an ERD-based attribute reduction method(abbreviated as ERD-RM),and then combing with a specific example and the ERD's characteristics we discuss the differences and relationships between ERD-RM and existing reduction methods.Finally,combing with the four data sets in UCI,we compare the ERD-RM with existing methods through experimental analysis,and indicate the feasibility,effectiveness and characteristics of ERD-RM.The theoretical analysis and practical applications show that ERD has good structural features and interpretability and can integrate decision preference into the roughness measure system in a straightforward manner.Therefore,it can not only enrich the existing theories,but also has wide application value in many fields,such as inductive learning,artificial intelligence,information science,biological science and data mining,etc.