Rule Discovery Technique Using Genetic Programming Combined with Apriori Algorithm

Various techniques have been proposed for rule discovery using classification learning. In general, the learning speed of a system using genetic programming (GP) [1] is slow. However, a learning system which can acquire higher-order knowledge by adjusting to the environment can be constructed, because the structure is treated at the same time. On the other hand, there is the Apriori algorithm [2], a rule generating technique for large databases. This is an association rule algorithm. The Apriori algorithm uses two values for rule construction: a support value and a confidence value. Depending on the setting of each index threshold, the search space can be reduced, or the candidate number of association rules can be increased. However, experience is necessary for setting an effective threshold. Both techniques have advantages and disadvantages as above. In this paper, we propose a rule discovery technique for databases using genetic programming combined with the Apriori algorithm. By using the combined rule generation learning method, it is expected to construct a system which can search for flexible rules in large databases.