{"id":"https://openalex.org/W3120848961","doi":"https://doi.org/10.1109/ialp51396.2020.9310508","title":"Statistical Machine Translation Approach for Lexical Normalization on Indonesian Text","display_name":"Statistical Machine Translation Approach for Lexical Normalization on Indonesian Text","publication_year":2020,"publication_date":"2020-12-04","ids":{"openalex":"https://openalex.org/W3120848961","doi":"https://doi.org/10.1109/ialp51396.2020.9310508","mag":"3120848961"},"language":"en","primary_location":{"id":"doi:10.1109/ialp51396.2020.9310508","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ialp51396.2020.9310508","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2020 International Conference on Asian Language Processing (IALP)","raw_type":"proceedings-article"},"type":"article","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/A5045905279","display_name":"Ajmal Kurnia","orcid":null},"institutions":[{"id":"https://openalex.org/I29617571","display_name":"University of Indonesia","ror":"https://ror.org/0116zj450","country_code":"ID","type":"education","lineage":["https://openalex.org/I29617571"]}],"countries":["ID"],"is_corresponding":false,"raw_author_name":"Ajmal Kurnia","raw_affiliation_strings":["Faculty of Computer Science, Universitas Indonesia, Depok, Indonesia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Faculty of Computer Science, Universitas Indonesia, Depok, Indonesia","institution_ids":["https://openalex.org/I29617571"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5004596561","display_name":"Evi Yulianti","orcid":"https://orcid.org/0000-0003-1951-4696"},"institutions":[{"id":"https://openalex.org/I29617571","display_name":"University of Indonesia","ror":"https://ror.org/0116zj450","country_code":"ID","type":"education","lineage":["https://openalex.org/I29617571"]}],"countries":["ID"],"is_corresponding":false,"raw_author_name":"Evi Yulianti","raw_affiliation_strings":["Faculty of Computer Science, Universitas Indonesia, Depok, Indonesia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Faculty of Computer Science, Universitas Indonesia, Depok, Indonesia","institution_ids":["https://openalex.org/I29617571"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I29617571"],"apc_list":null,"apc_paid":null,"fwci":0.2709,"has_fulltext":false,"cited_by_count":5,"citation_normalized_percentile":{"value":0.66599413,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"288","last_page":"293"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":1.0,"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":1.0,"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":0.9997000098228455,"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/T11550","display_name":"Text and Document Classification Technologies","score":0.986299991607666,"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/computer-science","display_name":"Computer science","score":0.8335652947425842},{"id":"https://openalex.org/keywords/normalization","display_name":"Normalization (sociology)","score":0.8067837953567505},{"id":"https://openalex.org/keywords/machine-translation","display_name":"Machine translation","score":0.7963274717330933},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.7741211652755737},{"id":"https://openalex.org/keywords/phrase","display_name":"Phrase","score":0.7419812083244324},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.7061245441436768},{"id":"https://openalex.org/keywords/evaluation-of-machine-translation","display_name":"Evaluation of machine translation","score":0.6016998291015625},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.5826009511947632},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.3614237308502197},{"id":"https://openalex.org/keywords/example-based-machine-translation","display_name":"Example-based machine translation","score":0.33894479274749756},{"id":"https://openalex.org/keywords/machine-translation-software-usability","display_name":"Machine translation software usability","score":0.3126027584075928}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8335652947425842},{"id":"https://openalex.org/C136886441","wikidata":"https://www.wikidata.org/wiki/Q926129","display_name":"Normalization (sociology)","level":2,"score":0.8067837953567505},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.7963274717330933},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.7741211652755737},{"id":"https://openalex.org/C2776224158","wikidata":"https://www.wikidata.org/wiki/Q187931","display_name":"Phrase","level":2,"score":0.7419812083244324},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7061245441436768},{"id":"https://openalex.org/C135784402","wikidata":"https://www.wikidata.org/wiki/Q6958279","display_name":"Evaluation of machine translation","level":5,"score":0.6016998291015625},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.5826009511947632},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3614237308502197},{"id":"https://openalex.org/C24687705","wikidata":"https://www.wikidata.org/wiki/Q3753284","display_name":"Example-based machine translation","level":3,"score":0.33894479274749756},{"id":"https://openalex.org/C148526163","wikidata":"https://www.wikidata.org/wiki/Q6723733","display_name":"Machine translation software usability","level":4,"score":0.3126027584075928},{"id":"https://openalex.org/C19165224","wikidata":"https://www.wikidata.org/wiki/Q23404","display_name":"Anthropology","level":1,"score":0.0},{"id":"https://openalex.org/C144024400","wikidata":"https://www.wikidata.org/wiki/Q21201","display_name":"Sociology","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/ialp51396.2020.9310508","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ialp51396.2020.9310508","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2020 International Conference on Asian Language Processing (IALP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.8199999928474426,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":28,"referenced_works":["https://openalex.org/W182831726","https://openalex.org/W1521626219","https://openalex.org/W2000877723","https://openalex.org/W2124807415","https://openalex.org/W2134800885","https://openalex.org/W2146574666","https://openalex.org/W2156985047","https://openalex.org/W2163942301","https://openalex.org/W2164107060","https://openalex.org/W2251589680","https://openalex.org/W2403864882","https://openalex.org/W2419539795","https://openalex.org/W2599008208","https://openalex.org/W2737886250","https://openalex.org/W2763882345","https://openalex.org/W2774974668","https://openalex.org/W2900700367","https://openalex.org/W2911944143","https://openalex.org/W2912843381","https://openalex.org/W2973417162","https://openalex.org/W2981458766","https://openalex.org/W2983418531","https://openalex.org/W2986227781","https://openalex.org/W6679855610","https://openalex.org/W6713512117","https://openalex.org/W6717262007","https://openalex.org/W6741402472","https://openalex.org/W6758129091"],"related_works":["https://openalex.org/W3204448004","https://openalex.org/W2122287718","https://openalex.org/W2587602790","https://openalex.org/W2740094425","https://openalex.org/W2566847733","https://openalex.org/W193726211","https://openalex.org/W3196981096","https://openalex.org/W2532807140","https://openalex.org/W2962780935","https://openalex.org/W2131079305"],"abstract_inverted_index":{"Lexical":[0],"normalization":[1,34],"is":[2],"an":[3,51,104],"important":[4],"task":[5],"to":[6,54,63],"be":[7],"performed":[8],"on":[9,42,74],"noisy":[10],"data,":[11],"such":[12],"as":[13],"social":[14],"media":[15],"posts,":[16],"before":[17],"using":[18,38,103],"the":[19,26,39,65,71,100,115],"data":[20,59],"for":[21,33],"further":[22,113],"analysis.":[23],"We":[24,48],"examine":[25],"potential":[27],"of":[28,35,102,117],"Statistical":[29],"Machine":[30],"Translation":[31],"(SMT)":[32],"Indonesian":[36,122],"text":[37],"translation":[40],"unit":[41],"both":[43,75],"phrase":[44,76],"and":[45,60,77,90,108],"character":[46,78],"levels.":[47],"also":[49,98],"used":[50],"external":[52,105],"corpus":[53],"generate":[55],"additional":[56],"language":[57,106],"model":[58,107],"pre-normalization":[61,110],"rules":[62,111],"enhance":[64,114],"SMT":[66,72,118],"system.":[67],"The":[68],"result":[69],"shows":[70],"systems":[73,119],"levels":[79],"are":[80],"outperforming":[81],"various":[82],"baseline":[83],"in":[84,120],"Word":[85],"Error":[86],"Rate":[87],"(WER)":[88],"score":[89],"Bilingual":[91],"Understudy":[92],"Evaluation":[93],"(BLEU)":[94],"score.":[95],"This":[96],"research":[97],"demonstrates":[99],"effect":[101],"applying":[109],"can":[112],"effectiveness":[116],"normalizing":[121],"text.":[123]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1},{"year":2021,"cited_by_count":1}],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2025-10-10T00:00:00"}