{"id":"https://openalex.org/W4206952096","doi":"https://doi.org/10.1109/ssci50451.2021.9659923","title":"Improving transformer model translation for low resource South African languages using BERT","display_name":"Improving transformer model translation for low resource South African languages using BERT","publication_year":2021,"publication_date":"2021-12-05","ids":{"openalex":"https://openalex.org/W4206952096","doi":"https://doi.org/10.1109/ssci50451.2021.9659923"},"language":"en","primary_location":{"id":"doi:10.1109/ssci50451.2021.9659923","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ssci50451.2021.9659923","pdf_url":null,"source":{"id":"https://openalex.org/S4363604921","display_name":"2021 IEEE Symposium Series on Computational Intelligence (SSCI)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE Symposium Series on Computational Intelligence (SSCI)","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/A5013907436","display_name":"Paddington Chiguvare","orcid":null},"institutions":[{"id":"https://openalex.org/I69552723","display_name":"University of Pretoria","ror":"https://ror.org/00g0p6g84","country_code":"ZA","type":"education","lineage":["https://openalex.org/I69552723"]}],"countries":["ZA"],"is_corresponding":true,"raw_author_name":"Paddington Chiguvare","raw_affiliation_strings":["University of Pretoria, Pretoria, South Africa"],"affiliations":[{"raw_affiliation_string":"University of Pretoria, Pretoria, South Africa","institution_ids":["https://openalex.org/I69552723"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5063774442","display_name":"Christopher W. Cleghorn","orcid":"https://orcid.org/0000-0002-7860-0650"},"institutions":[{"id":"https://openalex.org/I192619145","display_name":"University of the Witwatersrand","ror":"https://ror.org/03rp50x72","country_code":"ZA","type":"education","lineage":["https://openalex.org/I192619145"]}],"countries":["ZA"],"is_corresponding":false,"raw_author_name":"Christopher W Cleghorn","raw_affiliation_strings":["School of Computer Science and Applied Mathematics, University of Witwatersrand, Johannesburg, South Africa"],"affiliations":[{"raw_affiliation_string":"School of Computer Science and Applied Mathematics, University of Witwatersrand, Johannesburg, South Africa","institution_ids":["https://openalex.org/I192619145"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5013907436"],"corresponding_institution_ids":["https://openalex.org/I69552723"],"apc_list":null,"apc_paid":null,"fwci":0.377,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.61777902,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"8"},"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":0.9993000030517578,"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/T13629","display_name":"Text Readability and Simplification","score":0.973800003528595,"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/transformer","display_name":"Transformer","score":0.7920074462890625},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7637678384780884},{"id":"https://openalex.org/keywords/zulu","display_name":"Zulu","score":0.7367669939994812},{"id":"https://openalex.org/keywords/machine-translation","display_name":"Machine translation","score":0.6940216422080994},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.6114306449890137},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5727909207344055},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.512269914150238},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.34749215841293335},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.1199025809764862},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.09690698981285095}],"concepts":[{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.7920074462890625},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7637678384780884},{"id":"https://openalex.org/C2778105672","wikidata":"https://www.wikidata.org/wiki/Q10179","display_name":"Zulu","level":2,"score":0.7367669939994812},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.6940216422080994},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.6114306449890137},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5727909207344055},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.512269914150238},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.34749215841293335},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.1199025809764862},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.09690698981285095},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/ssci50451.2021.9659923","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ssci50451.2021.9659923","pdf_url":null,"source":{"id":"https://openalex.org/S4363604921","display_name":"2021 IEEE Symposium Series on Computational Intelligence (SSCI)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE Symposium Series on Computational Intelligence (SSCI)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.44999998807907104,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":21,"referenced_works":["https://openalex.org/W2101105183","https://openalex.org/W2594661830","https://openalex.org/W2896457183","https://openalex.org/W2901616036","https://openalex.org/W2949303037","https://openalex.org/W2951165112","https://openalex.org/W2962784628","https://openalex.org/W2969696767","https://openalex.org/W2970929998","https://openalex.org/W3000562670","https://openalex.org/W3006381853","https://openalex.org/W3013746816","https://openalex.org/W3014432386","https://openalex.org/W3030384960","https://openalex.org/W3035390927","https://openalex.org/W4287780357","https://openalex.org/W4288323699","https://openalex.org/W4385245566","https://openalex.org/W6769263558","https://openalex.org/W6778689135","https://openalex.org/W6779829939"],"related_works":["https://openalex.org/W3176018525","https://openalex.org/W2903533908","https://openalex.org/W2775554247","https://openalex.org/W2903810591","https://openalex.org/W3026554633","https://openalex.org/W2888520903","https://openalex.org/W2890256614","https://openalex.org/W2949454572","https://openalex.org/W3066373881","https://openalex.org/W2970247882"],"abstract_inverted_index":{"Language":[0],"pre-training":[1],"methods":[2],"such":[3],"as":[4],"BERT":[5,100,196,225,246],"have":[6],"led":[7],"to":[8,130,161,180],"significant":[9],"performance":[10],"gains":[11],"for":[12,30,77,91,136,153,171,201],"a":[13,232],"wide":[14],"range":[15],"of":[16,61,190,231,239],"natural":[17],"language":[18,42,49,78,103,137],"processing":[19],"tasks.":[20],"This":[21],"paper":[22],"investigates":[23],"the":[24,62,74,88,116,126,133,147,175,195,205,224,237,250],"improvement":[25,219],"in":[26,41,236],"transformer":[27,75,117,122,145,165,242],"model":[28,118,123,166,226,243,247],"translation":[29,124,203],"low":[31,47,63,182],"resource":[32,48,64,183],"South":[33,45,65,184],"African":[34,46,66,185],"languages":[35,230],"when":[36,223],"pre-trained":[37,55],"models":[38,76,135],"are":[39],"used":[40],"translation.":[43],"A":[44],"multilingual":[50,99],"BERT,":[51],"dubbed":[52],"SALR-mBERT":[53,68,86,97,134,152,168],"is":[54,193,199,208,227,253],"from":[56],"scratch":[57],"using":[58,229],"all":[59],"nine":[60],"languages.":[67,186],"performs":[69],"better":[70],"than":[71,83,107,141,213,255],"mBERT":[72,131],"and":[73,95,113,132,159,210,217],"corpora":[79,104,138],"with":[80,105,139],"approximately":[81,214],"less":[82,106,212],"600000":[84,108,142,215,256],"sentences.":[85,143,257],"obtained":[87,125,146],"best":[89,127,148],"results":[90,128,149],"Sesotho,":[92],"Ndebele,":[93],"Swati":[94],"Venda.":[96],"outperforms":[98,115,167],"(mBERT)":[101],"on":[102],"sentences":[109],"by":[110,119,151,169],"about":[111],"29%":[112],"also":[114],"50%.":[120],"The":[121,144,164,187],"compared":[129,179],"more":[140],"followed":[150],"Northern":[154],"Sotho,":[155],"Tsonga,":[156],"Xhosa,":[157],"Tswana":[158],"Zulu":[160,172],"English":[162],"translations.":[163],"15%":[170],"which":[173],"has":[174,211],"largest":[176],"corpus":[177,206,251],"size":[178,207,252],"other":[181],"main":[188],"finding":[189],"this":[191],"work":[192],"that":[194],"type":[197],"module":[198],"beneficial":[200],"machine":[202],"if":[204,249],"small":[209],"sentences,":[216],"further":[218],"can":[220],"be":[221],"gained":[222],"trained":[228],"similar":[233],"nature":[234],"like":[235],"case":[238],"SALR-mBERT.":[240],"However,":[241],"translations":[244,248],"outperform":[245],"greater":[254]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}