{"id":"https://openalex.org/W2008244516","doi":"https://doi.org/10.1109/icde.2012.111","title":"On Text Clustering with Side Information","display_name":"On Text Clustering with Side Information","publication_year":2012,"publication_date":"2012-04-01","ids":{"openalex":"https://openalex.org/W2008244516","doi":"https://doi.org/10.1109/icde.2012.111","mag":"2008244516"},"language":"en","primary_location":{"id":"doi:10.1109/icde.2012.111","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icde.2012.111","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2012 IEEE 28th International Conference on Data Engineering","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5028089542","display_name":"Char\u0173 C. Aggarwal","orcid":"https://orcid.org/0000-0003-2579-7581"},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]},{"id":"https://openalex.org/I4210114115","display_name":"IBM Research - Thomas J. Watson Research Center","ror":"https://ror.org/0265w5591","country_code":"US","type":"facility","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210114115"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Charu C. Aggarwal","raw_affiliation_strings":["IBM Thomas J. Watson Research Center, Hawthorne, NY, USA","IBM T.J. Watson Research Center, Hawthorne, NY, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Thomas J. Watson Research Center, Hawthorne, NY, USA","institution_ids":["https://openalex.org/I4210114115"]},{"raw_affiliation_string":"IBM T.J. Watson Research Center, Hawthorne, NY, USA","institution_ids":["https://openalex.org/I1341412227"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087513337","display_name":"Yuchen Zhao","orcid":"https://orcid.org/0000-0003-4780-093X"},"institutions":[{"id":"https://openalex.org/I39422238","display_name":"University of Illinois Chicago","ror":"https://ror.org/02mpq6x41","country_code":"US","type":"education","lineage":["https://openalex.org/I39422238"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yuchen Zhao","raw_affiliation_strings":["University of Illinois, Chicago, Chicago, IL, USA","University of Illinois at Chicago Chicago, IL USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Illinois, Chicago, Chicago, IL, USA","institution_ids":["https://openalex.org/I39422238"]},{"raw_affiliation_string":"University of Illinois at Chicago Chicago, IL USA","institution_ids":["https://openalex.org/I39422238"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5036357902","display_name":"Philip S. Yu","orcid":"https://orcid.org/0000-0002-3491-5968"},"institutions":[{"id":"https://openalex.org/I39422238","display_name":"University of Illinois Chicago","ror":"https://ror.org/02mpq6x41","country_code":"US","type":"education","lineage":["https://openalex.org/I39422238"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Philip S. Yu","raw_affiliation_strings":["University of Illinois, Chicago, Chicago, IL, USA","University of Illinois at Chicago Chicago, IL USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Illinois, Chicago, Chicago, IL, USA","institution_ids":["https://openalex.org/I39422238"]},{"raw_affiliation_string":"University of Illinois at Chicago Chicago, IL USA","institution_ids":["https://openalex.org/I39422238"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":35,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"894","last_page":"904"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11106","display_name":"Data Management and Algorithms","score":0.9987000226974487,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11106","display_name":"Data Management and Algorithms","score":0.9987000226974487,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10637","display_name":"Advanced Clustering Algorithms Research","score":0.9983999729156494,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10064","display_name":"Complex Network Analysis Techniques","score":0.9975000023841858,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/cluster-analysis","display_name":"Cluster analysis","score":0.8760428428649902},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8210103511810303},{"id":"https://openalex.org/keywords/data-mining","display_name":"Data mining","score":0.5693747401237488},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5226709246635437},{"id":"https://openalex.org/keywords/document-clustering","display_name":"Document clustering","score":0.5218952298164368},{"id":"https://openalex.org/keywords/probabilistic-logic","display_name":"Probabilistic logic","score":0.48951470851898193},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.48784446716308594},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.4457387626171112},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.2747257351875305},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.26065677404403687}],"concepts":[{"id":"https://openalex.org/C73555534","wikidata":"https://www.wikidata.org/wiki/Q622825","display_name":"Cluster analysis","level":2,"score":0.8760428428649902},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8210103511810303},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.5693747401237488},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5226709246635437},{"id":"https://openalex.org/C177937566","wikidata":"https://www.wikidata.org/wiki/Q4223102","display_name":"Document clustering","level":3,"score":0.5218952298164368},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.48951470851898193},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.48784446716308594},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.4457387626171112},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2747257351875305},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.26065677404403687},{"id":"https://openalex.org/C94625758","wikidata":"https://www.wikidata.org/wiki/Q7163","display_name":"Politics","level":2,"score":0.0},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.0},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/icde.2012.111","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icde.2012.111","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2012 IEEE 28th International Conference on Data Engineering","raw_type":"proceedings-article"},{"id":"pmh:oai:CiteSeerX.psu:10.1.1.297.7974","is_oa":false,"landing_page_url":"http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.297.7974","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"http://charuaggarwal.net/text-icde.pdf","raw_type":"text"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":46,"referenced_works":["https://openalex.org/W33276683","https://openalex.org/W170573592","https://openalex.org/W316065036","https://openalex.org/W631140850","https://openalex.org/W1509894598","https://openalex.org/W1533841329","https://openalex.org/W1575476631","https://openalex.org/W1651093245","https://openalex.org/W1925932413","https://openalex.org/W1956559956","https://openalex.org/W1958077162","https://openalex.org/W1971784203","https://openalex.org/W1996510517","https://openalex.org/W1996764654","https://openalex.org/W2013029404","https://openalex.org/W2014615172","https://openalex.org/W2040058125","https://openalex.org/W2057712948","https://openalex.org/W2092957858","https://openalex.org/W2095897464","https://openalex.org/W2106545428","https://openalex.org/W2116237179","https://openalex.org/W2123549998","https://openalex.org/W2126626732","https://openalex.org/W2133576408","https://openalex.org/W2140809065","https://openalex.org/W2150461699","https://openalex.org/W2165515835","https://openalex.org/W2166914830","https://openalex.org/W2167036382","https://openalex.org/W2216446631","https://openalex.org/W2319660501","https://openalex.org/W2325227998","https://openalex.org/W2434205482","https://openalex.org/W3011290429","https://openalex.org/W4231029117","https://openalex.org/W4236122429","https://openalex.org/W4241122026","https://openalex.org/W4250619709","https://openalex.org/W4255387915","https://openalex.org/W4285719527","https://openalex.org/W6637231022","https://openalex.org/W6640308466","https://openalex.org/W6677504887","https://openalex.org/W6678337573","https://openalex.org/W6684570415"],"related_works":["https://openalex.org/W2804364458","https://openalex.org/W1795405792","https://openalex.org/W2019737068","https://openalex.org/W2899601636","https://openalex.org/W4254379378","https://openalex.org/W3015674157","https://openalex.org/W4206655101","https://openalex.org/W52853789","https://openalex.org/W4237592971","https://openalex.org/W2387982377"],"abstract_inverted_index":{"Text":[0],"clustering":[1,101,135,165,199],"has":[2],"become":[3],"an":[4,183,197,221],"increasingly":[5],"important":[6],"problem":[7],"in":[8,22,25,47,72,193,212],"recent":[9],"years":[10],"because":[11,137],"of":[12,16,52,65,98,107,118,144,208,218],"the":[13,30,41,58,70,73,88,104,119,134,142,145,154,164,171,216],"tremendous":[14,96],"amount":[15,97],"unstructured":[17],"data":[18,42,210],"which":[19,84,185],"is":[20,43,54,121],"available":[21,46,55],"various":[23],"forms":[24],"online":[26],"forums":[27],"such":[28,68,124,220],"as":[29,69,168],"web,":[31],"social":[32],"networks,":[33],"and":[34],"other":[35,81],"information":[36,99,120],"networks.":[37],"In":[38,123,178],"most":[39],"cases,":[40,125],"not":[44],"purely":[45],"text":[48,59,89],"form.":[49],"A":[50],"lot":[51],"side-information":[53,62,109,132],"along":[56],"with":[57,190],"documents.":[60],"Such":[61,91],"may":[63,93,110],"be":[64,111,128],"different":[66],"kinds,":[67],"links":[71],"document,":[74],"user-access":[75],"behavior":[76],"from":[77,173],"web":[78],"logs,":[79],"or":[80,149],"non-textual":[82],"attributes":[83,92],"are":[85],"embedded":[86],"into":[87,133],"document.":[90],"contain":[94],"a":[95,159,206],"for":[100,147],"purposes.":[102],"However,":[103],"relative":[105],"importance":[106],"this":[108,175,179],"difficult":[112],"to":[113,130,153,162,169,195,214],"estimate,":[114],"especially":[115],"when":[116],"some":[117],"noisy.":[122],"it":[126,138],"can":[127,139,150],"risky":[129],"incorporate":[131],"process,":[136,166],"either":[140],"improve":[141],"quality":[143],"representation":[146],"clustering,":[148],"add":[151],"noise":[152],"process.":[155],"Therefore,":[156],"we":[157,181],"need":[158],"principled":[160],"way":[161],"perform":[163],"so":[167],"maximize":[170],"advantages":[172,217],"using":[174,219],"side":[176],"information.":[177],"paper,":[180],"design":[182],"algorithm":[184],"combines":[186],"classical":[187],"partitioning":[188],"algorithms":[189],"probabilistic":[191],"models":[192],"order":[194,213],"create":[196],"effective":[198],"approach.":[200,222],"We":[201],"present":[202],"experimental":[203],"results":[204],"on":[205],"number":[207],"real":[209],"sets":[211],"illustrate":[215]},"counts_by_year":[{"year":2020,"cited_by_count":2},{"year":2019,"cited_by_count":1},{"year":2018,"cited_by_count":6},{"year":2017,"cited_by_count":2},{"year":2016,"cited_by_count":6},{"year":2015,"cited_by_count":8},{"year":2014,"cited_by_count":3},{"year":2013,"cited_by_count":5},{"year":2012,"cited_by_count":2}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}