Data Mining is the process of obtaining high level knowledge by automatically discovering information from data in the form of rules and patterns. Data mining seeks to discover knowledge that is accurate, comprehensible and interesting. Association rule mining is a well established method of data mining that identifies significant correlations between items in transactional data. Measures like support count, comprehensibility and interestingness, used for evaluating a rule can be thought of as different objectives of association rule mining problem. In the thesis work we solve the association rule-mining problem using genetic algorithm. In the present work, we use the random sampling method. A perfect sample will improve the correctness of the rules generated by the algorithm. We will test in the approach only with the numerical valued attributes and must test with the categorical attributes also.
Loading....