Mining Partial Periodic Patterns in Time Series Data

Speaker: CAO Huiping

Abstract

The search for mining weak periodic patterns in time series data is an active topic. Given the fact that rarely a real world dataset is perfectly periodic, the coming presentation will introduce some efficient approaches for mining partial periodic patterns.

Given some period length, several algorithms for finding partial periodic patterns, by exploring some interesting properties related to partial periodicity, such as the Apriori property and the max-subpattern hit set peroperty, are presented. These algorithms needs only two scans over the time series data. Further, some method also are presebted to discover approximate and partial periodicities, when no period length is known in advance.

Referred Papers