Mining Interesting Periodicities of Temporal Patterns

Anjana Kakoti Mahanta, Hung Son Nguyen.

Data mining also known as knowledge discovery from datasets has been recognized as an important area of database research. This area can be defined as efficiently discovering interesting patterns from large data sets. In this paper a generic method has been proposed to extract interesting periodicities of patterns from large datasets where the transactions in the data sets are associated with patterns and time intervals in which the patterns hold. Considering the hierarchy associated with time stamps of the form day-date-hour-minutes-seconds, different types of periodic patterns such as daily, weekly, monthly patterns can be extracted.

PDF full paper