Abstract- Association rule mining plays a very important role in various data mining process. The fundamental idea of association rules is to mine the interesting (positive) frequent patterns from a transaction database. The point of this study is to build up the new model for mining interesting negative and positive association rules out of a transactional data set. For the mining of rule mining a variety of algorithm are used such as Apriori algorithm and multiple level minimum support based algorithm. Some algorithm is wonder performance but produces negative association rule and also has the problem of multi-scan database. MLMS-GA association rule mining based algorithm is proposed. In this mthod a multi-level multiple support of data table as 0 and 1 is used. The divided process reduces the scanning time of database and get reduced set of association rules.

Keywords: Data Mining, Association Rule Mining, Positive Rule Mining, Negative Rule Mining, Genetic Algorithm, Apriori Algorithm.