Tree based data fusion approach for mining temporal patterns

Discovering time profiled temporal patterns from time stamped transaction datasets is addressed in our previous research works which includes proposing new support estimation techniques, similarity measures for computing similarity between temporal patterns. This paper proposes a novel approach for discovering temporal pattern by introducing the concept of data fusion w.r.t the temporal pattern tree. The tree is generated for each timeslot and then the trees obtained for individual timeslots are merged or fused to get the overall tree for the entire dataset. The concept of tree based data fusion helps to prune elements efficiently and well ahead during pattern mining process. A pruning function is also introduced in this paper to prune invalid temporal patterns.