A hybrid Bees Swarm Optimization and Tabu Search algorithm for Association rule mining

The current world wide web is featured by big volumes of data. The classical association rules mining algorithms dealt with data sets somehow in an efficient way and in reasonable time. However they are not capable to cope with a huge amount of data in the web context where the respond time must be very short. In this paper a new hybrid algorithm called (HBSO-TS) is proposed for association rule mining. It is based on two well known meta-heuristics, which are Bees Swarm Optimization (BSO) and Tabu Search (TS). BSO is chosen for its remarkable diversification process while tabu search for its efficient intensification strategy. To make the idea simpler, BSO will browse the search space in such a way to cover most of its regions and the local exploration of each bee is computed by tabu search. The experimental study showed that, due to the simultaneous management of the two meta-heuristic in HBSO-TS, the empirical parameters setting becomes a difficult task. Despite this issue, results show that HBSO-TS algorithm outperforms the Association rule mining algorithms based on evolutionary computation and already proposed in the literature. In particular, we observed that the developed approach yields useful association rules in a reasonable time when comparing it with previous works.