{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T23:13:35Z","timestamp":1781046815370,"version":"3.54.1"},"reference-count":30,"publisher":"SAGE Publications","issue":"3","license":[{"start":{"date-parts":[[2026,4,5]],"date-time":"2026-04-05T00:00:00Z","timestamp":1775347200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"},{"start":{"date-parts":[[2026,4,5]],"date-time":"2026-04-05T00:00:00Z","timestamp":1775347200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"funder":[{"DOI":"10.13039\/501100001659","name":"Deutsche Forschungsgemeinschaft","doi-asserted-by":"publisher","award":["251654672"],"award-info":[{"award-number":["251654672"]}],"id":[{"id":"10.13039\/501100001659","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001659","name":"Deutsche Forschungsgemeinschaft","doi-asserted-by":"publisher","award":["410883423"],"award-info":[{"award-number":["410883423"]}],"id":[{"id":"10.13039\/501100001659","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["Information Visualization"],"published-print":{"date-parts":[[2026,7]]},"abstract":"<jats:p>Uniform Manifold Approximation and Projection (UMAP) has become a ubiquitous tool for high-dimensional data visualization, yet its interpretation is often hindered by the \u201ccartographic fallacy\u201d\u2014a cognitive bias where the embedding layout is assumed to be a faithful map of the data\u2019s intrinsic geometry, leading users to mistake algorithmic side-effects for genuine data properties. These artifacts stem not only from stochastic optimization but also from inherent mathematical assumptions regarding simplicial approximation and metric normalization. In this work, we present an interactive study aimed at diagnosing these mechanisms. We introduce a classification system derived from a suite of 10 synthetic 3D \u201cprobe\u201d datasets, categorizing distortions into spatial logic failures, topological loss, and metric distortion. Furthermore, we demonstrate a human-in-the-loop framework that pairs layout steering with parameter tuning to correct optimization traps and reveal topological trade-offs. This approach transforms UMAP, which is in its current form a static black box, into an explorable educational instrument, helping practitioners distinguish between genuine data features and algorithmic artifacts.<\/jats:p>","DOI":"10.1177\/14738716261434908","type":"journal-article","created":{"date-parts":[[2026,4,6]],"date-time":"2026-04-06T04:54:32Z","timestamp":1775451272000},"page":"358-372","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":1,"title":["Demystifying UMAP artifacts: An interactive study on diagnosis and steering using 3D probes"],"prefix":"10.1177","volume":"25","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-5047-214X","authenticated-orcid":false,"given":"Bin","family":"Chen","sequence":"first","affiliation":[{"name":"University of Konstanz, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8195-517X","authenticated-orcid":false,"given":"Yumeng","family":"Xue","sequence":"additional","affiliation":[{"name":"University of Konstanz, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1315-4602","authenticated-orcid":false,"given":"Patrick","family":"Paetzold","sequence":"additional","affiliation":[{"name":"University of Konstanz, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5803-2185","authenticated-orcid":false,"given":"Oliver","family":"Deussen","sequence":"additional","affiliation":[{"name":"University of Konstanz, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"179","published-online":{"date-parts":[[2026,4,5]]},"reference":[{"key":"e_1_3_2_2_2","doi-asserted-by":"crossref","unstructured":"Sedlmair M Munzner T Tory M. Empirical guidance on scatterplot and dimension reduction technique choices. IEEE Trans Vis Comput Graph 2013; 19(12). 2634\u20132643. https:\/\/doi.org\/10.1109\/TVCG.2013.153","DOI":"10.1109\/TVCG.2013.153"},{"key":"e_1_3_2_3_2","doi-asserted-by":"crossref","unstructured":"Sacha D Zhang L Sedlmair M et al. Visual interaction with dimensionality reduction: a structured literature analysis. IEEE Trans Vis Comput Graph 2017; 23(1). 241\u2013250. https:\/\/doi.org\/10.1109\/TVCG.2016.2598495","DOI":"10.1109\/TVCG.2016.2598495"},{"key":"e_1_3_2_4_2","doi-asserted-by":"crossref","unstructured":"McInnes L Healy J Melville J. UMAP: uniform manifold approximation and projection for dimension reduction. arXiv preprint 2018; https:\/\/doi.org\/10.48550\/arXiv.1802.03426","DOI":"10.32614\/CRAN.package.uwot"},{"key":"e_1_3_2_5_2","first-page":"2579","article-title":"Visualizing data using t-SNE","volume":"9","author":"van der Maaten L","year":"2008","unstructured":"van der Maaten L, Hinton G. Visualizing data using t-SNE. J Mach Learn Res 2008; 9: 2579\u20132605.","journal-title":"J Mach Learn Res"},{"key":"e_1_3_2_6_2","doi-asserted-by":"crossref","unstructured":"Becht E McInnes L Healy J et al. Dimensionality reduction for visualizing single-cell data using UMAP. Nat Biotechnol 2019; 37(1). 38\u201344. https:\/\/doi.org\/10.1038\/nbt.4314","DOI":"10.1038\/nbt.4314"},{"key":"e_1_3_2_7_2","doi-asserted-by":"crossref","unstructured":"Carter S Armstrong Z Schubert L et al. Exploring neural networks with activation atlases. Distill 2019; 4. https:\/\/doi.org\/10.23915\/distill.00015","DOI":"10.23915\/distill.00015"},{"key":"e_1_3_2_8_2","doi-asserted-by":"crossref","unstructured":"Wattenberg M Vi\u00e9gas F Johnson I. How to use t-SNE effectively. Distill 2016; 1(10). https:\/\/doi.org\/10.23915\/distill.00002","DOI":"10.23915\/distill.00002"},{"key":"e_1_3_2_9_2","unstructured":"Jeon H Park J Shin S et al. Stop misusing t-SNE and UMAP for visual analytics. arXiv preprint 2025."},{"key":"e_1_3_2_10_2","doi-asserted-by":"crossref","unstructured":"Chari T Pachter L. The specious art of single-cell genomics. PLoS Comput Biol 2023; 19(8). 1\u201320. https:\/\/doi.org\/10.1371\/journal.pcbi.1011288","DOI":"10.1371\/journal.pcbi.1011288"},{"key":"e_1_3_2_11_2","doi-asserted-by":"crossref","unstructured":"Kobak D Berens P. The art of using t-SNE for single-cell transcriptomics. Nat Commun 2019; 10(1). 5416. https:\/\/doi.org\/10.1038\/s41467-019-13056-x","DOI":"10.1038\/s41467-019-13056-x"},{"key":"e_1_3_2_12_2","doi-asserted-by":"crossref","unstructured":"Kobak D Linderman GC. Initialization is critical for preserving global data structure in both t-SNE and UMAP. Nat Biotechnol 2021; 39(2). 156\u2013157. https:\/\/doi.org\/10.1038\/s41587-020-00809-z","DOI":"10.1038\/s41587-020-00809-z"},{"key":"e_1_3_2_13_2","first-page":"1","article-title":"Understanding how dimension reduction tools work: an empirical approach to deciphering t-SNE, UMAP, TriMAP, and PaCMAP for data visualization","volume":"22","author":"Wang Y","year":"2021","unstructured":"Wang Y, Huang H, Rudin C, et al. Understanding how dimension reduction tools work: an empirical approach to deciphering t-SNE, UMAP, TriMAP, and PaCMAP for data visualization. J Mach Learn Res 2021; 22: 1\u201373.","journal-title":"J Mach Learn Res"},{"key":"e_1_3_2_14_2","first-page":"451","article-title":"Information retrieval perspective to nonlinear dimensionality reduction for data visualization","volume":"11","author":"Venna J","year":"2010","unstructured":"Venna J, Kaski S. Information retrieval perspective to nonlinear dimensionality reduction for data visualization. J Mach Learn Res 2010; 11: 451\u2013490.","journal-title":"J Mach Learn Res"},{"key":"e_1_3_2_15_2","doi-asserted-by":"crossref","unstructured":"Shepard RN. The analysis of proximities: multidimensional scaling with an unknown distance function. Psychometrika 1962; 27(2). 125\u2013140. https:\/\/doi.org\/10.1007\/BF02289630","DOI":"10.1007\/BF02289630"},{"key":"e_1_3_2_16_2","doi-asserted-by":"crossref","unstructured":"Lespinats S Aupetit M. Checkviz: sanity check and topological clues for linear and non-linear mappings. Computer Graphics Forum 2011; 30(1). 113\u2013125. https:\/\/doi.org\/10.1111\/j.1467-8659.2010.01835.x","DOI":"10.1111\/j.1467-8659.2010.01835.x"},{"key":"e_1_3_2_17_2","doi-asserted-by":"crossref","unstructured":"Bian R Xue Y Zhou L et al. Implicit multidimensional projection of local subspaces. IEEE Trans Vis Comput Graph 2021; 27(2). 1558\u20131568. https:\/\/doi.org\/10.1109\/TVCG.2020.3030368","DOI":"10.1109\/TVCG.2020.3030368"},{"key":"e_1_3_2_18_2","doi-asserted-by":"crossref","unstructured":"Cooley SM Hamilton T Ray J et al. A novel metric reveals previously unrecognized distortion in dimensionality reduction of scRNA-seq data. bioRxiv 2020; https:\/\/doi.org\/10.1101\/689851.","DOI":"10.1101\/689851"},{"key":"e_1_3_2_19_2","doi-asserted-by":"crossref","unstructured":"Paul R Chalup SK. A study on validating non-linear dimensionality reduction using persistent homology. Pattern Recognit Lett 2017; 100: 160\u2013166. https:\/\/doi.org\/10.1016\/j.patrec.2017.09.032","DOI":"10.1016\/j.patrec.2017.09.032"},{"key":"e_1_3_2_20_2","doi-asserted-by":"crossref","unstructured":"Sankaran K Zhang S Chenab et al. Interactive visualization of metric distortion in nonlinear data embeddings using the distortions package. bioRxiv 2025. https:\/\/doi.org\/10.1101\/2025.08.21.671523.","DOI":"10.1101\/2025.08.21.671523"},{"key":"e_1_3_2_21_2","doi-asserted-by":"crossref","unstructured":"Wang M Wenskovitch J House L et al. Bridging cognitive gaps between user and model in interactive dimension reduction. Vis Inform 2021; 5(2). 13\u201325. https:\/\/doi.org\/10.1016\/j.visinf.2021.03.002","DOI":"10.1016\/j.visinf.2021.03.002"},{"key":"e_1_3_2_22_2","doi-asserted-by":"crossref","unstructured":"Endert A Fiaux P North C. Semantic interaction for visual text analytics. In: Proceedings of the SIGCHI conference on human factors in computing systems CHI \u201912 2012 pp.473\u2013482. Association for Computing Machinery. https:\/\/doi.org\/10.1145\/2207676.2207741.","DOI":"10.1145\/2207676.2207741"},{"key":"e_1_3_2_23_2","doi-asserted-by":"crossref","unstructured":"Brown ET Liu J Brodley CE et al. Dis-function: learning distance functions interactively. In: 2012 IEEE conference on visual analytics science and technology (VAST) 2012 pp.83\u201392. https:\/\/doi.org\/10.1109\/VAST.2012.6400483.","DOI":"10.1109\/VAST.2012.6400486"},{"key":"e_1_3_2_24_2","unstructured":"Self JZ Hu X House L et al. Designing usable interactive visual analytics tools for dimension reduction. In: CHI 2016 workshop on human-centered machine learning (HCML) 2016 p.7."},{"key":"e_1_3_2_25_2","doi-asserted-by":"crossref","unstructured":"Kwon BC Kim H Wall E et al. Axisketcher: interactive nonlinear axis mapping of visualizations through user drawings. IEEE Trans Vis Comput Graph 2017; 23(1). 221\u2013230. https:\/\/doi.org\/10.1109\/TVCG.2016.2598446","DOI":"10.1109\/TVCG.2016.2598446"},{"key":"e_1_3_2_26_2","doi-asserted-by":"crossref","unstructured":"Pezzotti N Lelieveldt BPF Van Der Maaten L et al. Approximated and user steerable tSNE for progressive visual analytics. IEEE Trans Vis Comput Graph 2017; 23(7). 1739\u20131752. https:\/\/doi.org\/10.1109\/TVCG.2016.2570755","DOI":"10.1109\/TVCG.2016.2570755"},{"key":"e_1_3_2_27_2","doi-asserted-by":"crossref","unstructured":"Wang ZJ Hohman F Chau DH. Wizmap: scalable interactive visualization for exploring large machine learning embeddings. arXiv preprint 2023. https:\/\/doi.org\/10.48550\/arXiv.2306.09328.","DOI":"10.18653\/v1\/2023.acl-demo.50"},{"key":"e_1_3_2_28_2","doi-asserted-by":"crossref","unstructured":"Sainburg T McInnes L Gentner TQ. Parametric UMAP embeddings for representation and semi-supervised learning. Neural Comput 2021; 33(11). 2881\u20132907. https:\/\/doi.org\/10.1162\/neco_a_01434","DOI":"10.1162\/neco_a_01434"},{"key":"e_1_3_2_29_2","doi-asserted-by":"crossref","unstructured":"Xia J Huang L Lin W et al. Interactive visual cluster analysis by contrastive dimensionality reduction. IEEE Trans Vis Comput Graph 2023; 29(1). 734\u2013744. https:\/\/doi.org\/10.1109\/TVCG.2022.3209423","DOI":"10.1109\/TVCG.2022.3209423"},{"key":"e_1_3_2_30_2","volume-title":"Computational topology: an introduction","author":"Edelsbrunner H","year":"2010","unstructured":"Edelsbrunner H, Harer J. Computational topology: an introduction. American Mathematical Society, 2010."},{"key":"e_1_3_2_31_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41587-020-00801-7"}],"container-title":["Information Visualization"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/14738716261434908","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/14738716261434908","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/14738716261434908","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T11:03:44Z","timestamp":1781003024000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/14738716261434908"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,5]]},"references-count":30,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026,7]]}},"alternative-id":["10.1177\/14738716261434908"],"URL":"https:\/\/doi.org\/10.1177\/14738716261434908","relation":{},"ISSN":["1473-8716","1473-8724"],"issn-type":[{"value":"1473-8716","type":"print"},{"value":"1473-8724","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,5]]}}}