Mining inter-transaction association rules from multiple time-series data

Association rule mining is one of the most widely used methods for discovering interesting relations between variables. Time series as a common sequence data have some unique character, such as pervasively connected, endless and time-related. Therefore research on multivariate time series data mining is a hot spot in data mining. This paper first compresses the continuous time series. Then in order to make the mining rules reflect the characteristics of multivariate time series data, our paper designs a new algorithm called IAMTL, which can mine the rules from the fix time span. For the reason that time series data have the characteristic of continuity, so an increment version of IATML is provided. At last, we use prerequisite and the consequent windows to verify the correctness of the rules.