A Novel Mining Algorithm for Periodic Clustering Sequential Patterns

In knowledge discovery, data mining of time series information has many important applications. Especially, sequential patterns and periodic patterns, which evolved from the association rule, have been applied in many useful practices. This paper presents another useful concept, the periodic clustering sequential (PCS) pattern, which uses clustering to mine valuable information from temporal or serially ordered data in a period of time. In traditional clustering algorithms, there was little discussion in time series. However, in many applications we must consider the time factor. For example, one can cluster patients according to symptoms of the illness under study, but this may just result in several clusters with specific symptoms for analyzing the distribution of patients. Adding time series analysis to the above investigation, we can examine the distribution of patients over the same or different seasons. For policymakers, the PCS pattern is more useful than traditional clustering result and provides a more effective support of decision-making. In our proposed algorithm, we employ two effective structures, a Time Cube and a Periodic Table. A Time Cube stores clusters that are grouped by time to measure the confidence of the PCS pattern. A Periodic Table is used to produce candidates for the PCS pattern. Using these two sound structures with our efficient algorithm, we can find the PCS pattern in an effective way.