{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/time-series-clustering-via-community","title":"Time Series Clustering via Community Detection in Networks","arxiv_id":"1508.04757","date":"2015-08-19","proceeding":null,"authors":["Leonardo N. Ferreira","Liang Zhao"],"abstract":"In this paper, we propose a technique for time series clustering using\ncommunity detection in complex networks. Firstly, we present a method to\ntransform a set of time series into a network using different distance\nfunctions, where each time series is represented by a vertex and the most\nsimilar ones are connected. Then, we apply community detection algorithms to\nidentify groups of strongly connected vertices (called a community) and,\nconsequently, identify time series clusters. Still in this paper, we make a\ncomprehensive analysis on the influence of various combinations of time series\ndistance functions, network generation methods and community detection\ntechniques on clustering results. Experimental study shows that the proposed\nnetwork-based approach achieves better results than various classic or\nup-to-date clustering techniques under consideration. Statistical tests confirm\nthat the proposed method outperforms some classic clustering algorithms, such\nas $k$-medoids, diana, median-linkage and centroid-linkage in various data\nsets. Interestingly, the proposed method can effectively detect shape patterns\npresented in time series due to the topological structure of the underlying\nnetwork constructed in the clustering process. At the same time, other\ntechniques fail to identify such patterns. Moreover, the proposed method is\nrobust enough to group time series presenting similar pattern but with time\nshifts and/or amplitude variations. In summary, the main point of the proposed\nmethod is the transformation of time series from time-space domain to\ntopological domain. Therefore, we hope that our approach contributes not only\nfor time series clustering, but also for general time series analysis tasks.","url_abs":"http://arxiv.org/abs/1508.04757v1","url_pdf":"http://arxiv.org/pdf/1508.04757v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"time-series-clustering-via-community","repo_url":"https://github.com/lnferreira/time_series_clustering_via_community_detection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"community-detection","task_name":"Community Detection"},{"task_slug":"time-series-1","task_name":"Time Series"},{"task_slug":"time-series","task_name":"Time Series Analysis"},{"task_slug":"time-series-clustering","task_name":"Time Series Clustering"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}