{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2023,8,22]],"date-time":"2023-08-22T04:56:24Z","timestamp":1692680184816},"reference-count":67,"publisher":"Wiley","issue":"3","license":[{"start":{"date-parts":[[2023,4,16]],"date-time":"2023-04-16T00:00:00Z","timestamp":1681603200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":["wires.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["WIREs Data Min & Knowl"],"published-print":{"date-parts":[[2023,5]]},"abstract":"Abstract<\/jats:title>Outlier or anomaly detection is an important task in data analysis. We discuss the problem from a geometrical perspective and provide a framework which exploits the metric structure of a data set. Our approach rests on the manifold assumption<\/jats:italic>, that is, that the observed, nominally high\u2010dimensional data lie on a much lower dimensional manifold and that this intrinsic structure can be inferred with manifold learning methods. We show that exploiting this structure significantly improves the detection of outlying observations in high dimensional data. We also suggest a novel, mathematically precise and widely applicable distinction between distributional<\/jats:italic> and structural<\/jats:italic> outliers based on the geometry and topology of the data manifold that clarifies conceptual ambiguities prevalent throughout the literature. Our experiments focus on functional data as one class of structured high\u2010dimensional data, but the framework we propose is completely general and we include image and graph data applications. Our results show that the outlier structure of high\u2010dimensional and non\u2010tabular data can be detected and visualized using manifold learning methods and quantified using standard outlier scoring methods applied to the manifold embedding vectors.<\/jats:p>This article is categorized under:\nTechnologies > Structure Discovery and Clustering<\/jats:p><\/jats:list-item>\nFundamental Concepts of Data and Knowledge > Data Concepts<\/jats:p><\/jats:list-item>\nTechnologies > Visualization<\/jats:p><\/jats:list-item>\n<\/jats:list><\/jats:p>","DOI":"10.1002\/widm.1491","type":"journal-article","created":{"date-parts":[[2023,4,17]],"date-time":"2023-04-17T02:25:23Z","timestamp":1681698323000},"update-policy":"http:\/\/dx.doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["A geometric framework for outlier detection in high\u2010dimensional data"],"prefix":"10.1002","volume":"13","author":[{"ORCID":"http:\/\/orcid.org\/0000-0002-4893-5812","authenticated-orcid":false,"given":"Moritz","family":"Herrmann","sequence":"first","affiliation":[{"name":"Department of Statistics Ludwig Maximilians University Munich Germany"}]},{"given":"Florian","family":"Pfisterer","sequence":"additional","affiliation":[{"name":"Department of Statistics Ludwig Maximilians University Munich Germany"}]},{"given":"Fabian","family":"Scheipl","sequence":"additional","affiliation":[{"name":"Department of Statistics Ludwig Maximilians University Munich Germany"}]}],"member":"311","published-online":{"date-parts":[[2023,4,16]]},"reference":[{"key":"e_1_2_13_2_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-47578-3"},{"key":"e_1_2_13_3_1","doi-asserted-by":"publisher","DOI":"10.1145\/376284.375668"},{"key":"e_1_2_13_4_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00371-019-01673-y"},{"key":"e_1_2_13_5_1","doi-asserted-by":"publisher","DOI":"10.1136\/practneurol-2017-001719"},{"key":"e_1_2_13_6_1","first-page":"2","article-title":"The irises of the gasp\u00e9 peninsula","volume":"59","author":"Anderson E.","year":"1935","journal-title":"Bulletin of the American Iris Society"},{"key":"e_1_2_13_7_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-018-24874-2"},{"key":"e_1_2_13_8_1","doi-asserted-by":"publisher","DOI":"10.1080\/00401706.1983.10487840"},{"key":"e_1_2_13_9_1","doi-asserted-by":"publisher","DOI":"10.1162\/089976603321780317"},{"key":"e_1_2_13_10_1","doi-asserted-by":"publisher","DOI":"10.1007\/3-540-49257-7_15"},{"key":"e_1_2_13_11_1","doi-asserted-by":"publisher","DOI":"10.1145\/335191.335388"},{"key":"e_1_2_13_12_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10618-015-0444-8"},{"key":"e_1_2_13_13_1","first-page":"659","volume-title":"Proceedings of the sixteenth international conference on artificial intelligence and statistics","author":"Cl\u00e9men\u00e7on S.","year":"2013"},{"key":"e_1_2_13_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.acha.2006.04.006"},{"key":"e_1_2_13_15_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-33037-0_14"},{"key":"e_1_2_13_16_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.csda.2020.106960"},{"key":"e_1_2_13_17_1","doi-asserted-by":"publisher","DOI":"10.1109\/JAS.2019.1911747"},{"key":"e_1_2_13_18_1","volume-title":"UCI machine learning repository","author":"Dua D.","year":"2017"},{"key":"e_1_2_13_19_1","doi-asserted-by":"publisher","DOI":"10.18637\/jss.v051.i04"},{"key":"e_1_2_13_20_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1469-1809.1936.tb02137.x"},{"key":"e_1_2_13_21_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2012.05.002"},{"key":"e_1_2_13_22_1","doi-asserted-by":"publisher","DOI":"10.1161\/01.CIR.101.23.e215"},{"key":"e_1_2_13_23_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0152173"},{"key":"e_1_2_13_24_1","unstructured":"Guan S. &Loew M.(2021).A novel intrinsic measure of data separability. arXiv:2109.05180 [Cs Math Stat].http:\/\/arxiv.org\/abs\/2109.05180"},{"key":"e_1_2_13_25_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-44781-0_28"},{"key":"e_1_2_13_26_1","doi-asserted-by":"publisher","DOI":"10.3390\/stats4040057"},{"key":"e_1_2_13_27_1","first-page":"171","volume-title":"The 2nd International Conference on e\u2010Health and Telemedicine (ICEHTM\u20102014), 5","author":"Isenkul M.","year":"2014"},{"key":"e_1_2_13_28_1","doi-asserted-by":"publisher","DOI":"10.1142\/S0219649220400134"},{"key":"e_1_2_13_29_1","doi-asserted-by":"publisher","DOI":"10.1080\/10618600.2020.1807353"},{"key":"e_1_2_13_30_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICDE.2012.88"},{"key":"e_1_2_13_31_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-0-387-39351-3"},{"key":"e_1_2_13_32_1","doi-asserted-by":"publisher","DOI":"10.1080\/1351847X.2019.1647864"},{"key":"e_1_2_13_33_1","doi-asserted-by":"publisher","DOI":"10.1201\/b11431"},{"issue":"86","key":"e_1_2_13_34_1","first-page":"2579","article-title":"Visualizing data using t\u2010SNE","volume":"9","author":"Maaten L.","year":"2008","journal-title":"Journal of Machine Learning Research"},{"key":"e_1_2_13_35_1","doi-asserted-by":"publisher","DOI":"10.1145\/3394053"},{"key":"e_1_2_13_36_1","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1802.03426"},{"key":"e_1_2_13_37_1","unstructured":"McInnes L. Healy J. &Melville J.(2020).UMAP: Uniform manifold approximation and projection for dimension reduction. arXiv:1802.03426 [Cs Stat].http:\/\/arxiv.org\/abs\/1802.03426"},{"issue":"12","key":"e_1_2_13_38_1","first-page":"411","article-title":"Dimensionality estimation, manifold learning and function approximation using tensor voting","volume":"11","author":"Mordohai P.","year":"2010","journal-title":"Journal of Machine Learning Research"},{"key":"e_1_2_13_39_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-30463-0_27"},{"key":"e_1_2_13_40_1","unstructured":"Nane S. Nayar S. &Murase H.(1996).Columbia object image library: COIL\u201020. Dept. Comp. Sci. Columbia University New York Tech. Rep."},{"key":"e_1_2_13_41_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11749-020-00750-y"},{"key":"e_1_2_13_42_1","doi-asserted-by":"publisher","DOI":"10.1137\/090762932"},{"key":"e_1_2_13_43_1","doi-asserted-by":"publisher","DOI":"10.1145\/3219819.3220042"},{"key":"e_1_2_13_44_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0020-0190(99)00156-8"},{"key":"e_1_2_13_45_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0304-4149(97)00028-8"},{"key":"e_1_2_13_46_1","doi-asserted-by":"publisher","DOI":"10.1002\/0470013192.bsa239"},{"key":"e_1_2_13_47_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10994-008-5093-3"},{"key":"e_1_2_13_48_1","doi-asserted-by":"publisher","DOI":"10.1093\/biomet\/asx012"},{"key":"e_1_2_13_49_1","doi-asserted-by":"publisher","DOI":"10.1093\/biomet\/asv021"},{"key":"e_1_2_13_50_1","volume-title":"Robust regression and outlier detection","author":"Rousseeuw P. J.","year":"2005"},{"key":"e_1_2_13_51_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.290.5500.2323"},{"issue":"24","key":"e_1_2_13_52_1","first-page":"665","article-title":"Learning minimum volume sets","volume":"7","author":"Scott C. D.","year":"2006","journal-title":"The Journal of Machine Learning Research"},{"key":"e_1_2_13_53_1","doi-asserted-by":"crossref","unstructured":"Souvenir R. &Pless R.(2005).Manifold clustering. Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1 648\u2013653.https:\/\/doi.org\/10.1109\/ICCV.2005.149","DOI":"10.1109\/ICCV.2005.149"},{"key":"e_1_2_13_54_1","doi-asserted-by":"publisher","DOI":"10.1111\/biom.13706"},{"key":"e_1_2_13_55_1","doi-asserted-by":"publisher","DOI":"10.1117\/12.148698"},{"key":"e_1_2_13_56_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.290.5500.2319"},{"key":"e_1_2_13_57_1","doi-asserted-by":"publisher","DOI":"10.1186\/s40537-020-00320-x"},{"key":"e_1_2_13_58_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-13062-5_20"},{"key":"e_1_2_13_59_1","doi-asserted-by":"publisher","DOI":"10.1007\/BF02288916"},{"key":"e_1_2_13_60_1","doi-asserted-by":"publisher","DOI":"10.1080\/10618600.2019.1575226"},{"key":"e_1_2_13_61_1","doi-asserted-by":"publisher","DOI":"10.1109\/INFVIS.2005.1532142"},{"key":"e_1_2_13_62_1","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.2016.1256813"},{"key":"e_1_2_13_63_1","doi-asserted-by":"publisher","DOI":"10.2991\/ijcis.11.1.50"},{"key":"e_1_2_13_64_1","doi-asserted-by":"publisher","DOI":"10.1007\/BF02287916"},{"key":"e_1_2_13_65_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICC.2006.255127"},{"key":"e_1_2_13_66_1","doi-asserted-by":"publisher","DOI":"10.1002\/widm.1280"},{"key":"e_1_2_13_67_1","doi-asserted-by":"publisher","DOI":"10.1002\/sam.11161"},{"key":"e_1_2_13_68_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10994-013-5334-y"}],"container-title":["WIREs Data Mining and Knowledge Discovery"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/wires.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/widm.1491","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,21]],"date-time":"2023-08-21T15:18:39Z","timestamp":1692631119000},"score":1,"resource":{"primary":{"URL":"https:\/\/wires.onlinelibrary.wiley.com\/doi\/10.1002\/widm.1491"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,4,16]]},"references-count":67,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2023,5]]}},"alternative-id":["10.1002\/widm.1491"],"URL":"https:\/\/doi.org\/10.1002\/widm.1491","archive":["Portico"],"relation":{},"ISSN":["1942-4787","1942-4795"],"issn-type":[{"value":"1942-4787","type":"print"},{"value":"1942-4795","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,4,16]]},"assertion":[{"value":"2022-04-14","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2022-12-08","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2023-04-16","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}