Genetic Algorithm And Its Application In Data Mining Research

With the development of computer and information technology, the increase rate of information has become exponential. 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 new technology that can eliminate the rude and retian the essential,the Data Mining technology which can pick the knowledge and information from the vast data then came into being.Data Mining(DM) is aprocess that pick previously unknown and potentially useful information and technology from large volumes of incomplete, fuzzy and stochastic data with noise. However for the inherent uncertainty and complex of data and the attributes,some methods show their inability in some cases.Soft computing is good at dealing with such dilemma,therefore it is valuable to study data mining techniques in the frame of soft computing.Genetic algorithm as a good search,evolutionary tool in data mining field has a unique charm. It simulates natural selection and biological genetic mechanism and generates the offspring by genetic operators.Through the iterativeness of population,the fitnesses of the individuals is improved and finally the individual with the highest fitness just is the optimal solution or suboptimal solution of the optimizationp roblem. In this paper,geneticalgorithm is presented to mine of association rules of data mining.The Primary research work of the Paper was as follows:1.This paper commented data mining,explains the definition of data mining,the objects of data mining,the main process of data mining,classification,main technology and so on.2.Based on the basic concepts of the association rules,this dissertation classifies and summarizes its species and value roundly.3.An important method--genetic algorithms(GA)in data mining is introduced,and the development,main theory,characteristic and application of an approach are recommended.Present a multi-population genetic algorithm to improve the problem which of premature convergence and inefficient loval search in the standard gentic algorithm.4.Discussed the application of genetic algorithms for data mining ideas and key issues, the improved genetic algorithm is applied to wine datasets which by UCI provide by the association rule mining, wine properties and quality attributes are hidden between the association, show that the algorithm is feasible nature, can be obtained from the database of rules with strong predictive ability.