{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,2,22]],"date-time":"2025-02-22T00:46:08Z","timestamp":1740185168042,"version":"3.37.3"},"reference-count":35,"publisher":"Oxford University Press (OUP)","issue":"20","license":[{"start":{"date-parts":[[2022,8,30]],"date-time":"2022-08-30T00:00:00Z","timestamp":1661817600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,10,14]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:sec>\n                  <jats:title>Motivation<\/jats:title>\n                  <jats:p>The medical data are complex in nature as terms that appear in records usually appear in different contexts. Through this article, we investigate various bio model\u2019s embeddings (BioBERT, BioELECTRA and PubMedBERT) on their understanding of \u2018negation and speculation context\u2019 wherein we found that these models were unable to differentiate \u2018negated context\u2019 versus \u2018non-negated context\u2019. To measure the understanding of models, we used cosine similarity scores of negated sentence embeddings versus non-negated sentence embeddings pairs. For improving these models, we introduce a generic super tuning approach to enhance the embeddings on \u2018negation and speculation context\u2019 by utilizing a synthesized dataset.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>After super-tuning the models, we can see that the model\u2019s embeddings are now understanding negative and speculative contexts much better. Furthermore, we fine-tuned the super-tuned models on various tasks and we found that the model has outperformed the previous models and achieved state-of-the-art on negation, speculation cue and scope detection tasks on BioScope abstracts and Sherlock dataset. We also confirmed that our approach had a very minimal trade-off in the performance of the model in other tasks like natural language inference after super-tuning.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Availability and implementation<\/jats:title>\n                  <jats:p>The source code, data and the models are available at: https:\/\/github.com\/comprehend\/engg-ai-research\/tree\/uncertainty-super-tuning.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btac593","type":"journal-article","created":{"date-parts":[[2022,8,30]],"date-time":"2022-08-30T13:26:02Z","timestamp":1661865962000},"page":"4790-4796","source":"Crossref","is-referenced-by-count":1,"title":["No means \u2018No\u2019: a non-improper modeling approach, with embedded speculative context"],"prefix":"10.1093","volume":"38","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4488-6053","authenticated-orcid":false,"given":"Priya","family":"Tiwary","sequence":"first","affiliation":[{"name":"Saama AI Research Lab , Pune 411057, India"}]},{"given":"Akshayraj","family":"Madhubalan","sequence":"additional","affiliation":[{"name":"Saama AI Research Lab , Pune 411057, India"}]},{"given":"Amit","family":"Gautam","sequence":"additional","affiliation":[{"name":"Saama AI Research Lab , Pune 411057, India"}]}],"member":"286","published-online":{"date-parts":[[2022,8,30]]},"reference":[{"year":"2012","author":"Abu-Jbara","key":"2022101415191712500_btac593-B1"},{"key":"2022101415191712500_btac593-B2","doi-asserted-by":"crossref","first-page":"696","DOI":"10.1136\/jamia.2010.003228","article-title":"Biomedical negation scope detection with conditional random fields","volume":"17","author":"Agarwal","year":"2010","journal-title":"J. Am. Med. Inform. Assoc"},{"year":"2020","author":"Britto","key":"2022101415191712500_btac593-B3"},{"first-page":"169","year":"2018","author":"Cer","key":"2022101415191712500_btac593-B4"},{"key":"2022101415191712500_btac593-B5","doi-asserted-by":"crossref","first-page":"301","DOI":"10.1006\/jbin.2001.1029","article-title":"A simple algorithm for identifying negated findings and diseases in discharge summaries","volume":"34","author":"Chapman","year":"2001","journal-title":"J. Biomed. Inform"},{"key":"2022101415191712500_btac593-B6","doi-asserted-by":"crossref","first-page":"1","DOI":"10.5121\/ijaia.2019.10101","article-title":"Attention-based deep learning system for negation and assertion detection in clinical notes","volume":"10","author":"Chen","year":"2019","journal-title":"IJAIA"},{"first-page":"340","year":"2012","author":"Chowdhury","key":"2022101415191712500_btac593-B7"},{"year":"2020","author":"Clark","key":"2022101415191712500_btac593-B8"},{"year":"1999","author":"Collier","key":"2022101415191712500_btac593-B9"},{"first-page":"51","year":"2010","author":"Councill","key":"2022101415191712500_btac593-B10"},{"year":"2018","author":"Devlin","key":"2022101415191712500_btac593-B11"},{"first-page":"495","year":"2016","author":"Fancellu","key":"2022101415191712500_btac593-B12"},{"first-page":"58","year":"2017","author":"Fancellu","key":"2022101415191712500_btac593-B13"},{"key":"2022101415191712500_btac593-B14","article-title":"Neural networks for cross-lingual negation scope detection","author":"Fancellu","year":"2018","journal-title":"Clin. Orthop. Relat. Res. (CoRR)"},{"key":"2022101415191712500_btac593-B15","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3458754","article-title":"Domain-specific language model pretraining for biomedical natural language processing","volume":"3","author":"Gu","year":"2022","journal-title":"ACM Trans. Comput. Healthcare"},{"first-page":"143","year":"2021","author":"Kanakarajan","key":"2022101415191712500_btac593-B16"},{"first-page":"5739","year":"2020","author":"Khandelwal","key":"2022101415191712500_btac593-B17"},{"first-page":"319","year":"2012","author":"Lapponi","key":"2022101415191712500_btac593-B18"},{"key":"2022101415191712500_btac593-B19","first-page":"1367","article-title":"BioBERT: a pre-trained biomedical language representation model for biomedical text mining","author":"Lee","year":"2019","journal-title":"Bioinformatics"},{"year":"2018","author":"Li","key":"2022101415191712500_btac593-B20"},{"first-page":"1","year":"2016","author":"Mirsky","key":"2022101415191712500_btac593-B21"},{"year":"2009","author":"Morante","key":"2022101415191712500_btac593-B22"},{"key":"2022101415191712500_btac593-B23","doi-asserted-by":"crossref","first-page":"598","DOI":"10.1136\/jamia.2001.0080598","article-title":"Use of general-purpose negation detection to augment concept indexing of medical documents: a quantitative study using the UMLS","volume":"8","author":"Mutalik","year":"2001","journal-title":"J. Am. Med. Inform. Assoc"},{"year":"2014","author":"Packard","key":"2022101415191712500_btac593-B24"},{"key":"2022101415191712500_btac593-B25","article-title":"Negbio: a high-performance tool for negation and uncertainty detection in radiology reports","author":"Peng","year":"2017","journal-title":"AMIA Jt Summits Transl Sci Proc."},{"year":"2016","author":"Qian","key":"2022101415191712500_btac593-B26"},{"key":"2022101415191712500_btac593-B27","first-page":"1","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","volume":"21","author":"Raffel","year":"2020","journal-title":"J. Mach. Learn. Res"},{"first-page":"310","year":"2012","author":"Read","key":"2022101415191712500_btac593-B28"},{"year":"2019","author":"Reimers","key":"2022101415191712500_btac593-B29"},{"first-page":"1586","year":"2019","author":"Shivade","key":"2022101415191712500_btac593-B30"},{"key":"2022101415191712500_btac593-B31","doi-asserted-by":"crossref","first-page":"i49","DOI":"10.1093\/bioinformatics\/btx238","article-title":"BIOSSES: a semantic sentence similarity estimation system for the biomedical domain","volume":"33","author":"So\u011fanc\u0131o\u011flu","year":"2017","journal-title":"Bioinformatics"},{"volume-title":"Proceedings of the Workshop on Current Trends in Biomedical Natural Language Processing, Columbus, Ohio. Association for Computational Linguistics","year":"2008","author":"Szarvas","key":"2022101415191712500_btac593-B32"},{"key":"2022101415191712500_btac593-B33","first-page":"6000","article-title":"Attention is all you need","author":"Vaswani","year":"2017"},{"first-page":"335","year":"2012","author":"White","key":"2022101415191712500_btac593-B34"},{"year":"2019","author":"Zhang","key":"2022101415191712500_btac593-B35"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/advance-article-pdf\/doi\/10.1093\/bioinformatics\/btac593\/46105615\/btac593.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/38\/20\/4790\/46535041\/btac593.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/38\/20\/4790\/46535041\/btac593.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,10,14]],"date-time":"2022-10-14T21:22:56Z","timestamp":1665782576000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/38\/20\/4790\/6678982"}},"subtitle":[],"editor":[{"given":"Inanc","family":"Birol","sequence":"additional","affiliation":[]}],"short-title":[],"issued":{"date-parts":[[2022,8,30]]},"references-count":35,"journal-issue":{"issue":"20","published-online":{"date-parts":[[2022,8,30]]},"published-print":{"date-parts":[[2022,10,14]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btac593","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"type":"print","value":"1367-4803"},{"type":"electronic","value":"1367-4811"}],"subject":[],"published-other":{"date-parts":[[2022,10,15]]},"published":{"date-parts":[[2022,8,30]]}}}