An extended two-phase architecture for mining time series data

Abstract

Time series data vary with time. In the past, most of the researches focused on the matching of feature points or measuring of the similarities. They could successfully represent the feature patterns in a visualized way. In the mean while, those researches did not sufficiently describe the results in simple and understandable words. In this research, a two-phase architecture for mining time series data is introduced. By combining some different mining techniques, the difficulties mentioned above may be overcome. This architecture mainly consists of Exploratory Data Analysis (EDA) and techniques related to mining association rules. After the phase I analysis, quantitative association rules are obtained by phase II. Meanwhile, the rules of the architecture are able to be verified by accuracy analysis. Finally, a result of comparison with the traditional data mining techniques and this architecture shows that the two-phase architecture is superior to traditional techniques to the time series data.