Periodic-Pattern Tree Miner: An Efficient Algorithm to Mine the Periodic Patterns From the Spatio-Temporal Database
When analyzing many applications that track object movements, it can be seen that, many movements obey periodic patterns. For example, people tend to wake up at the same time and follow more or less the same path to their work each day. Discovering the hidden periodic patterns from a spatio-temporal database, if there are any, could unveil important information to a data analyst as well as facilitate data management substantially. The problem of discovering periodic patterns from historical object movements is very challenging. Existing methods for finding periodic patterns are quite time and memory consuming. In this paper, the authors define the problem of mining periodic patterns from spatiotemporal database and propose an algorithm called, periodic-pattern tree miner that is used for finding all the periodic patterns.