Incremental Association Rule Mining Through Vertical Transaction ID

Association rule mining is a popular data mining technique which gives us valuable relationships among different items in a dataset. In dynamic databases, new transactions are appended as time advances. This may introduce new association rules and some existing association rules would become invalid. Thus, the maintenance of association rules for dynamic databases is an important problem. Several incremental algorithms, is proposed to deal with this problem. In this paper we proposed algorithm VTII (Vertical Transaction Id Intersections). This algorithm reduces a number of times to scan the database (old and new) to generate frequent pattern. As a result, the algorithm has execution time faster than that of previous Algorithms. This paper also conducts experiments to show the performance of the proposed algorithm. The result shows that the proposed algorithm has a good performance. Keywords—Association rule, Dynamic maintenance, Incremental , Vertical.