Font Size: a A A

Research On The Method Of Intelligent Data Analysis Based On Rough Set And Concept Lattice

Posted on:2005-02-20Degree:DoctorType:Dissertation
Country:ChinaCandidate:J Y LiangFull Text:PDF
GTID:1118360185996953Subject:Computer software and theory
Abstract/Summary:PDF Full Text Request
Rough set theory and concept lattice theory proposed separately by Poland scholar Z. Pawlak and Germany scholar R.Wille are powerful tool for data analysis. They have gained increasingly studying in recent years, and have been successfully used in widely field such as machine learning, decision analysis, process control, knowledge discovery in database, expert system etc. Therefore, studying the basic theory of rough set and concept lattice and applying it in knowledge discovery have important significance.In this thesis, we make a systematic and in-depth investigation on the basic theory of rough set and concept lattice and the method of knowledge discovery.In aspect of characterization of measures for rough set data analysis, a concept called the inclusion degree is introduced and the relationship between the rough inclusion and the inclusion degree is analyzed. Several important relations between inclusion degree and measures on rough set data analysis are established. We also validated that some measures of rough set data analysis can come down to inclusion degree. These results will be very helpful for people to understand the essence of rough set data analysis, and lie the foundation of measures defined for rough set data analysis.In aspect of information measure in information systems, a new type of information entropy and rough entropy are proposed, and the relation between Shannon entropy and rough entropy, and the relation between information entropy and knowledge granulation are established. In incomplete information systems, the concepts of granulation measure, information entropy, rough entropy and knowledge granulation are proposed, and properties of these concepts are given, the relations of these concepts are established. These results will be very helpful for understanding the essence of concept approximation and establishing granular computing in incomplete information systems.In aspect of extension of rough set theory in incomplete information systems, we bring forward a variable precision rough set model, and demonstrate its mathematics properties, witch can go on with knowledge reduction effectively on different conceptual level through the introduction of fuzzy technology. The results provided a new method for solving the data processing question of incomplete information systems by using rough set theory.
Keywords/Search Tags:Rough set, concept lattice, information systems, data analysis, incomplete information systems, inclusion degree, information entropy, rough entropy, knowledge granulation, measure, knowledge reduction, algebraic system, rule extracting
PDF Full Text Request
Related items