Based On Rough Set Theory Data Mining Technology And Its Application Of Potential Consumers Of Private Cars

As the database technology mature gradually, the rapid popularization of data applications, and the rapid development of the Internet, the accumulation of data quantity human beings are exponential growth in speed. Based on the database of knowledge discovery is developed in recent years of a new technology, it processing database of large amounts of data, dig out the deeper knowledge, more decision making of information. Data mining is from a lot of, not completely, noise, random, fuzzy data, extract implicit in them, people don’t know in advance, value of knowledge. The data pretreatment is the key step of data mining and data mining on the necessary work before work, in this stage of the attributes reduction and standardization of data processing.Rough set theory is the1980s by polish mathematician Pawlak first proposed, it is a kind of deal with fuzzy and uncertainty knowledge mathematical tools. One attribute reduction is the core content of rough set theory, attribute reduction is to keep the classification ability under the premise of constant, delete redundant attributes. The traditional attribute reduction algorithm is based on the algorithm of main storage. Based on relational database operation of the attribute reduction algorithm using relation database of operators and SQL itself operation, more efficient than traditional algorithm to data mining. In a relational database how to attribute reduction, to get the most efficient data mining, this article is the key.er according to a rough set theory based on the method of the data mining, to market survey data as the research object, by the appropriate reduction algorithm, and concludes that the simplified decision-making sets, the analysis private cars potential consumers group characteristics. Puts forward a kind of heuristic reduction algorithm, in the attribute reduction first before defines a condition attribute classification ability intensity, according to each attribute ability intensity of the decision table sort, and then used the method of judgment ideally suited to manufacture progre ive-scan attributes as unnecessary whether attribute, solve the attribute reduction may appear DuoGe reduction results to need to rely on people choose satisfactory results.I according to paper design process, using Delphi design procedures and make use of the SQL Sever2000data storage, eventually completed the attribute reduction, value reduction for the programming of the module, and thus more intuitive show the practice process and the results of the study.