E-MAP: Efficiently Mining Asynchronous Periodic Patterns

Mining periodic patterns in temporal dataset plays an important role in data mining and knowledge discovery tasks. In this paper, we propose a novel algorithm E-MAP (Efficient Mining of Asynchronous Periodic Patterns) for efficient mining of asynchronous periodic patterns in large temporal datasets. Our proposed algorithm discovers all maximal complex patterns in a single step and single scan without mining single event and multi events patterns. To check the effectiveness of our approach, we also provide detailed experimental results on real and artificial large temporal datasets. Our experimental results suggest that mining asynchronous periodic patterns using our proposed algorithm is fast and efficient than as compared to previous approach SMCA, which is a three-step based algorithm for mining maximal complex patterns and requires depth-firstenumeration for mining multi events and maximal complex patterns.