{"id":"https://openalex.org/W2251589680","doi":"https://doi.org/10.18653/v1/w15-4305","title":"A Normalizer for UGC in Brazilian Portuguese","display_name":"A Normalizer for UGC in Brazilian Portuguese","publication_year":2015,"publication_date":"2015-01-01","ids":{"openalex":"https://openalex.org/W2251589680","doi":"https://doi.org/10.18653/v1/w15-4305","mag":"2251589680"},"language":"en","primary_location":{"id":"doi:10.18653/v1/w15-4305","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/w15-4305","pdf_url":"https://www.aclweb.org/anthology/W15-4305.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Workshop on Noisy User-generated Text","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.aclweb.org/anthology/W15-4305.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5032431924","display_name":"Magali Sanches Duran","orcid":"https://orcid.org/0000-0002-3843-4600"},"institutions":[{"id":"https://openalex.org/I4210131883","display_name":"Brazilian Society of Computational and Applied Mathematics","ror":"https://ror.org/03kcw4w74","country_code":"BR","type":"other","lineage":["https://openalex.org/I4210131883"]}],"countries":["BR"],"is_corresponding":false,"raw_author_name":"Magali Sanches Duran","raw_affiliation_strings":["NILC -Center for Computational Linguistics So Paulo University (USP) So Carlos-SP, Brazil","NILC -Center for Computational Linguistics S\u00e3o Paulo University (USP) S\u00e3o Carlos-SP, Brazil"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NILC -Center for Computational Linguistics So Paulo University (USP) So Carlos-SP, Brazil","institution_ids":["https://openalex.org/I4210131883"]},{"raw_affiliation_string":"NILC -Center for Computational Linguistics S\u00e3o Paulo University (USP) S\u00e3o Carlos-SP, Brazil","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075011291","display_name":"Maria das Gra\u00e7as Volpe Nunes","orcid":"https://orcid.org/0000-0002-2776-6140"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Maria das Gra\u00e7as Volpe Nunes","raw_affiliation_strings":["NILC-Center for Computational Linguistics S\u00e3o Paulo University (USP) S\u00e3o Carlos-SP, Brazil"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NILC-Center for Computational Linguistics S\u00e3o Paulo University (USP) S\u00e3o Carlos-SP, Brazil","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5071517439","display_name":"Lucas Avan\u00e7o","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lucas Avan\u00e7o","raw_affiliation_strings":["NILC -Center for Computational Linguistics S\u00e3o Paulo University (USP) S\u00e3o Carlos-SP, Brazil"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NILC -Center for Computational Linguistics S\u00e3o Paulo University (USP) S\u00e3o Carlos-SP, Brazil","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.0288,"has_fulltext":true,"cited_by_count":9,"citation_normalized_percentile":{"value":0.78827382,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"38","last_page":"47"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9975000023841858,"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"}},"topics":[{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9975000023841858,"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/T10028","display_name":"Topic Modeling","score":0.9965999722480774,"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/T13083","display_name":"Advanced Text Analysis Techniques","score":0.9898999929428101,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/portuguese","display_name":"Portuguese","score":0.715377926826477},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.7121821045875549},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7039333581924438},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6562861800193787},{"id":"https://openalex.org/keywords/lexicon","display_name":"Lexicon","score":0.6173968315124512},{"id":"https://openalex.org/keywords/normalization","display_name":"Normalization (sociology)","score":0.6141395568847656},{"id":"https://openalex.org/keywords/sentence","display_name":"Sentence","score":0.5785123705863953},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5491332411766052},{"id":"https://openalex.org/keywords/arabic","display_name":"Arabic","score":0.4914132356643677},{"id":"https://openalex.org/keywords/f1-score","display_name":"F1 score","score":0.46966469287872314},{"id":"https://openalex.org/keywords/brazilian-portuguese","display_name":"Brazilian Portuguese","score":0.4656418561935425},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.3179629445075989}],"concepts":[{"id":"https://openalex.org/C35219183","wikidata":"https://www.wikidata.org/wiki/Q5146","display_name":"Portuguese","level":2,"score":0.715377926826477},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.7121821045875549},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7039333581924438},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6562861800193787},{"id":"https://openalex.org/C2778121359","wikidata":"https://www.wikidata.org/wiki/Q8096","display_name":"Lexicon","level":2,"score":0.6173968315124512},{"id":"https://openalex.org/C136886441","wikidata":"https://www.wikidata.org/wiki/Q926129","display_name":"Normalization (sociology)","level":2,"score":0.6141395568847656},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.5785123705863953},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5491332411766052},{"id":"https://openalex.org/C96455323","wikidata":"https://www.wikidata.org/wiki/Q13955","display_name":"Arabic","level":2,"score":0.4914132356643677},{"id":"https://openalex.org/C148524875","wikidata":"https://www.wikidata.org/wiki/Q6975395","display_name":"F1 score","level":2,"score":0.46966469287872314},{"id":"https://openalex.org/C2778880076","wikidata":"https://www.wikidata.org/wiki/Q750553","display_name":"Brazilian Portuguese","level":3,"score":0.4656418561935425},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.3179629445075989},{"id":"https://openalex.org/C144024400","wikidata":"https://www.wikidata.org/wiki/Q21201","display_name":"Sociology","level":0,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C19165224","wikidata":"https://www.wikidata.org/wiki/Q23404","display_name":"Anthropology","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.18653/v1/w15-4305","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/w15-4305","pdf_url":"https://www.aclweb.org/anthology/W15-4305.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Workshop on Noisy User-generated Text","raw_type":"proceedings-article"},{"id":"pmh:oai:www.producao.usp.br:BDPI/49416","is_oa":false,"landing_page_url":"http://www.producao.usp.br/handle/BDPI/49416","pdf_url":null,"source":{"id":"https://openalex.org/S4306401373","display_name":"Scientific Electronic Library Online (S\u00e3o Paulo Research Foundation, Latin American and Caribbean Center on Health Sciences Information, Conselho Nacional de Desenvolvimento Cient\u00edfico e Tecnol\u00f3gico)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I2982035970","host_organization_name":"Costa Rican Department of Social Security","host_organization_lineage":["https://openalex.org/I2982035970"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"conferenceObject"}],"best_oa_location":{"id":"doi:10.18653/v1/w15-4305","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/w15-4305","pdf_url":"https://www.aclweb.org/anthology/W15-4305.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Workshop on Noisy User-generated Text","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.7900000214576721,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[{"id":"https://openalex.org/G1987319257","display_name":null,"funder_award_id":"8.248/91","funder_id":"https://openalex.org/F4320319527","funder_display_name":"Samsung Eletr\u00f4nica da Amaz\u00f4nia"},{"id":"https://openalex.org/G2315157603","display_name":null,"funder_award_id":"8.248/91","funder_id":"https://openalex.org/F4320332195","funder_display_name":"Samsung"}],"funders":[{"id":"https://openalex.org/F4320319527","display_name":"Samsung Eletr\u00f4nica da Amaz\u00f4nia","ror":null},{"id":"https://openalex.org/F4320332195","display_name":"Samsung","ror":"https://ror.org/04w3jy968"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2251589680.pdf","grobid_xml":"https://content.openalex.org/works/W2251589680.grobid-xml"},"referenced_works_count":19,"referenced_works":["https://openalex.org/W63056884","https://openalex.org/W568396059","https://openalex.org/W1507805082","https://openalex.org/W1647671624","https://openalex.org/W1773803948","https://openalex.org/W1874835824","https://openalex.org/W2030841861","https://openalex.org/W2032355794","https://openalex.org/W2096438711","https://openalex.org/W2111216449","https://openalex.org/W2116533968","https://openalex.org/W2117995488","https://openalex.org/W2153804780","https://openalex.org/W2163942301","https://openalex.org/W2251592786","https://openalex.org/W2252107166","https://openalex.org/W2616272086","https://openalex.org/W3206634752","https://openalex.org/W6637852806"],"related_works":["https://openalex.org/W2399900321","https://openalex.org/W2789286307","https://openalex.org/W2899664355","https://openalex.org/W4319311403","https://openalex.org/W2306266016","https://openalex.org/W4254443907","https://openalex.org/W2886753397","https://openalex.org/W4377968335","https://openalex.org/W2544314260","https://openalex.org/W4387674232"],"abstract_inverted_index":{"User-generated":[0],"contents":[1],"(UGC)":[2],"represent":[3],"an":[4],"important":[5],"source":[6],"of":[7,18,32,40,44,85,88,110,118,123,131,146,153],"information":[8],"for":[9,30,51,61],"governments,":[10],"companies,":[11],"political":[12],"candidates":[13],"and":[14,24,29,35,46,75,103,142,158],"consumers.":[15],"However,":[16],"most":[17],"the":[19,111,116,151],"Natural":[20],"Language":[21],"Processing":[22],"tools":[23],"techniques":[25],"are":[26],"developed":[27],"from":[28,81,106,136,161],"texts":[31],"standard":[33],"language,":[34],"UGC":[36,62],"is":[37],"a":[38,58,66,68,72,82,86,129,144],"type":[39,117],"text":[41,119],"especially":[42],"full":[43],"creativity":[45],"idiosyncrasies,":[47],"which":[48,78,134,149],"represents":[49],"noise":[50],"NLP":[52],"purposes.":[53],"This":[54],"paper":[55],"presents":[56],"UGCNormal,":[57],"lexicon-based":[59],"tool":[60],"normalization.":[63],"It":[64],"encompasses":[65],"tokenizer,":[67],"sentence":[69],"segmentation":[70],"tool,":[71],"phonetic-based":[73],"speller":[74],"some":[76],"lexicons,":[77],"were":[79],"originated":[80],"deep":[83],"analysis":[84],"corpus":[87],"product":[89],"reviews":[90],"in":[91,98,128,140,143],"Brazilian":[92],"Portuguese.":[93],"The":[94,121],"normalizer":[95],"was":[96,125],"evaluated":[97],"two":[99],"different":[100],"data":[101],"sets":[102],"carried":[104],"out":[105],"31%":[107],"to":[108,138,163],"89%":[109],"appropriate":[112],"corrections,":[113],"depending":[114],"on":[115],"noise.":[120],"use":[122],"UGCNormal":[124],"also":[126],"validated":[127],"task":[130,145],"POS":[132],"tagging,":[133],"improved":[135,150],"91.35%":[137],"93.15%":[139],"accuracy":[141],"opinion":[147],"classification,":[148],"average":[152],"F1-score":[154,159],"measures":[155],"(F1-score":[156],"positive":[157],"negative)":[160],"0.736":[162],"0.758.":[164]},"counts_by_year":[{"year":2022,"cited_by_count":1},{"year":2020,"cited_by_count":1},{"year":2019,"cited_by_count":2},{"year":2018,"cited_by_count":1},{"year":2017,"cited_by_count":2},{"year":2016,"cited_by_count":2}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}