Abstract:

Association rules mining is one of the most studied subjects in data mining. In this work, new methods based on fuzzy logic and genetic algorithm, topics of soft computing and artificial intelligence, have been developed for quantitative association rules mining from databases that can have quantitative attributes. Especially, three different initial population methods that have been eliminated, the disadvantages of random initial population method in the first step of genetic algorithm have been used and the obtained results have been compared. To test the proposed methods, recordings of the Fırat University Electrical and Electronic Engineering students’ class grades have been selected as sample database and useful and interesting rules have effectively been mined

Keywords: