Generating Recommendations using an Association Rule Mining and Genetic Algorithm Combination

Abstract: A Recommender system is a subclass that seeks to find out the rating or preference that a user would give to an object of interest. The set of available recommender systems generate a recommendation based on historical information available. The information is based on a user`s taste, but not intend. Genetic recommendation offer real-time recommendations in a specific order, thereby overcoming drawbacks of existing systems. In this paper, an Association Rule Mining technique combining features of Eclat algorithm and Genetic Algorithm is proposed. The idea is to apply association rule mining technique Eclat for generating rules and further use genetic algorithm to optimize these rules. A performance comparison is done between results achieved by another popular Association Rule Algorithm, The Apriori algorithm and the results of Eclat-Genetic algorithm. It is observed experimentally that the Eclat-Genetic model gives 28.31 % better result than the existing Apriori algorithm in terms of accuracy.

Keywords: Recommender, agriculture, Eclat-Genetic, Association Rule Mining, rules.