The Application Of Genetic Algorithms In Data Mining

With the development of computer and information technology, the increase rate of information has become exponential. In the latest several decades, there were many super databases which spread all over the supermarket sell, bank saving, clerical work and scientific research. The tremendous increase of information made the traditional analytical method far from meeting the demands of society. How to find valuable information and knowledge from the vast data has become a very arduous task. People eagerly needed a technology that can eliminate the rude and retian the essential; meanwhile, can eliminate the false and retain the true. The Data Mining technology which can pick the knowledge and information from the vast data then came into being.Data Mining (DM) is a process that pick previously unknown and potentially useful information and technology from large volumes of incomplete, fuzzy and stochastic data with noise. It made use research achievements of many years in the areas of statistical and mathematical techniques, artificial intelligence and knowledge engineering etc. to form its theory system. It is a Cross-Discipline Field which integrated the technologys such as database, artificial intelligence, statistical and mathematical techniques, Description and Visualization, Parallel Computing etc..Data Mining is a Trans-discipline developed production impelled by the demands of applications. Furthermore, it is developed rapidly in recent years. The essential of the area is the combination of intelligence technology and database technology. It not only supplys the knowledge and strategy for the decider, but also brings economic benefit to the investor.Now there are many algorithmns applied in Data Mining. But many DM algorithmns are all involved in such problems as the incompleteness, imprecision and uncertainty of the data, that is, how to pick rules from inconsistent examples; the validity of rules, that is, how to deal with many useless rules found in DM; the choiceness of the rules, that is, now to make choice in the collision of rules; error control and algorithmn efficiency etc..