ODMC: Outlier detection on multivariate time series data based on clustering

Outlier detection on time series data plays an import role in life. In this paper we propose a method of outlier detection on time series data mainly aiming at the multivariate type. The improved ant colony algorithm is used for data clustering in the purpose of the classification of the time series data. Both the distance of inner-clusters and inter-clusters are considered to ensure the accuracy of the clustering. In addition we play an emphasis on the similarity between data points. The objects which have significant changes from the neighbors are identified as outliers. Experiments results show the algorithm is effective and efficient.