The Research Of Granularity Computing Based On Rough Set In Data Mining

With the extensive application of database technology, the amount of data in the database increases rapidly. Proposed the concept of Knowledge Discovery and Date Mining, in order to process the huge database in a higher level and find out laws and models to help people make better use of these data for decision-making. Knowledge Discovery is the non-trivial process that distinguish effective, innovative and potentially useful, and ultimately understandable pattern from the amount of data. Data mining is the process that extracts hidden, unknown and the potential value of information and knowledge form large amounts of data of the database. Data mining is the most critical steps in knowledge discovery, but also the technical difficulties in knowledge discovery, is the very active area of research nowadays.The theory of Rough Sets, presented by Polish mathematician Pawlak Z, is a powerful mathematical tool for analyzing uncertain, fuzzy knowledge. Rough sets, as a new hot spot in the field of artificial intelligence, can effectively deal with the expression and deduction of incomplete, uncertain knowledge. The theory of Rough Sets is specially fit for the application to Data-Mining because of it features. When thinking and solving problems, people choose "part then entirety",or "entirety then part", or alternative use of both, according to the need. The people can not only carry on the question in the different granularity world , moreover can jump very quickly from a granularity world to another granularity world, the round-trip freely, is not difficult. Therefore introduces the granularity concept in the data mining to have this very important significance.This paper will study granularity thought and applied it to data mining process, attributing reduction and extracting rules in the concept and granularity perspective, used in mining useful and interesting knowledge from large databases, which can be the solution of the problem that more data but less knowledge.Firstly,this paper summarized the data mining and the rough collection's correlation theories and the domestic and foreign research present situation,probed the technology hot spots of granularity computing and data mining and the trend development of both.Secondly,discussed reduction algorithm of rough set theory thoroughly, which included the attribute reduction and the attribute value reduction, In the present attribute reduction algorithm's foundation, this paper proposed one kind based on the conditional information entropy's attribute reduction improvement algorithm, simultaneously, applies the granularity thought in the rule extraction, proposed the rule extraction algorithm which based on the search granularity, from the top, establishes the multi-level granularity models. Finally aims at the tradition to have the impractical characteristic, proposed data mining model based on the rough set theory. This model including three modules:the data pretreatment, the attribute reduction and the rule extraction, then confirms this model's feasibility using the example.