Abstract: In almost every scientific field, measurements are performed over time. These observations lead to a collection of organized data called time series. The purpose of time-series data mining is to try to extract all meaningful knowledge from the shape of data. Even if humans have a natural capacity to perform these tasks, it remains a complex problem for computers. The first paper was based on study of various periodicity detection techniques and extracts their advantages and disadvantages. In this paper we intend to provide the result of new efficient technique for periodicity detection. This paper includes finding of three type of periodic pattern symbol periodicity, sequence periodicity or partial periodic pattern and segment or full-cycle periodic. The degrees of perfection calculated by confidence, and are mostly characterized by the presence of noise in the data. In this paper, we address the problem of detecting the periodicity rate of a time series database. Three types of periodicities are defined, and a scalable, computationally efficient algorithm is proposed for each type. The algorithms perform in O(nlogn) time for a time series of length n. Moreover, the proposed algorithms are extended in order to discover the periodic patterns of unknown periods at the same time without affecting the time complexity. Experimental results show that the proposed algorithms are highly accurate with respect to the discovered periodicity rates and periodic patterns. Real-data experiments demonstrate the practicality of the discovered periodic patterns.

Keywords: Time series, Data Mining, confidence, efficient algorithm, Periodicity detection.