An Algorithm for Time Series Data Mining Based on Clustering

This paper presents a new method for time series data mining. Discrete Fourier transform (DFT) is used to transform the time series data from time domain to frequency domain. By taking the transformed amplitude of power spectrum as the feature samples of the time series data, time series data can be mapped into a frequency domain space. We use OPTICS (ordering points to identify the cluster structure) algorithm to detect clusters in these data. Several simulations are given based on the price histories of California power market