Clustering Time Series by Network Community Analysis

Community analysis, a recently developed approach for partitioning the nodes of a network, is used as a tool for time series clustering. A network with N nodes is associated to the set of N time series, with the weight of the link (i, j) quantifying the similarity between the two corresponding time series. Then, searching for network communities allows one to identify groups of nodes (i.e., time series) with strong similarity. A quantitative assessment of the significance of the obtained partition is also provided.