Discovery of Temporal Association Rules in Multivariate Time Series

Abstract

This paper focuses on mining association rules in multivariate time series. Common association rule mining algorithms can only be applied to transactional data, and a typical application is market basket analysis. If we want to apply these algorithms on time-series data, changes need to be made. During temporal association rule mining, the natural temporal ordering of data and the temporal interval between the left and right patterns of a rule need to be considered. This paper reviews some methods for temporal association rule mining, and proposes two similar algorithms for the mining of frequent patterns in single and multivariate time series, both scalable and efficient. The pattern pruning and clustering is applied to reduce the number of patterns found. Temporal association rules are generated from the patterns found. Finally, the scalability and efficiency of the algorithms are demonstrated by evaluating it and comparing it to another similar work.