ETP-Mine: An Efficient Method for Mining Transitional Patterns
A Transaction database contains a set of transactions along with items and their associated timestamps. Transitional patterns are the patterns which specify the dynamic behavior of frequent patterns in a transaction database. To discover transitional patterns and their significant milestones, first the authors have to extract all frequent patterns and their supports using any frequent pattern generation algorithm. These frequent patterns are used in the generation of transitional patterns. The existing algorithm (TP-Mine) generates frequent patterns, some of which cannot be used in generation of transitional patterns. In this paper, they propose a modification to the existing algorithm, which prunes the candidate items to be used in the generation of frequent patterns.