{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,7,30]],"date-time":"2025-07-30T11:47:39Z","timestamp":1753876059918,"version":"3.41.2"},"reference-count":39,"publisher":"Oxford University Press (OUP)","license":[{"start":{"date-parts":[[2022,11,25]],"date-time":"2022-11-25T00:00:00Z","timestamp":1669334400000},"content-version":"vor","delay-in-days":328,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100004377","name":"Hong Kong Polytechnic University","doi-asserted-by":"publisher","award":["1-W182"],"award-info":[{"award-number":["1-W182"]}],"id":[{"id":"10.13039\/501100004377","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004377","name":"Hong Kong Polytechnic University","doi-asserted-by":"publisher","award":["G-YW4H"],"award-info":[{"award-number":["G-YW4H"]}],"id":[{"id":"10.13039\/501100004377","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,11,25]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>The Coronavirus Disease 2019 (COVID-19) pandemic has shifted the focus of research worldwide, and more than 10\u2009000 new articles per month have concentrated on COVID-19\u2013related topics. Considering this rapidly growing literature, the efficient and precise extraction of the main topics of COVID-19\u2013relevant articles is of great importance. The manual curation of this information for biomedical literature is labor-intensive and time-consuming, and as such the procedure is insufficient and difficult to maintain. In response to these complications, the BioCreative VII community has proposed a challenging task, LitCovid Track, calling for a global effort to automatically extract semantic topics for COVID-19 literature. This article describes our work on the BioCreative VII LitCovid Track. We proposed the LitCovid Ensemble Learning (LCEL) method for the tasks and integrated multiple biomedical pretrained models to address the COVID-19 multi-label classification problem. Specifically, seven different transformer-based pretrained models were ensembled for the initialization and fine-tuning processes independently. To enhance the representation abilities of the deep neural models, diverse additional biomedical knowledge was utilized to facilitate the fruitfulness of the semantic expressions. Simple yet effective data augmentation was also leveraged to address the learning deficiency during the training phase. In addition, given the imbalanced label distribution of the challenging task, a novel asymmetric loss function was applied to the LCEL model, which explicitly adjusted the negative\u2013positive importance by assigning different exponential decay factors and helped the model focus on the positive samples. After the training phase, an ensemble bagging strategy was adopted to merge the outputs from each model for final predictions. The experimental results show the effectiveness of our proposed approach, as LCEL obtains the state-of-the-art performance on the LitCovid dataset.<\/jats:p><jats:p>Database URL: https:\/\/github.com\/JHnlp\/LCEL<\/jats:p>","DOI":"10.1093\/database\/baac103","type":"journal-article","created":{"date-parts":[[2022,11,25]],"date-time":"2022-11-25T12:18:24Z","timestamp":1669378704000},"source":"Crossref","is-referenced-by-count":4,"title":["LitCovid ensemble learning for COVID-19 multi-label classification"],"prefix":"10.1093","volume":"2022","author":[{"given":"Jinghang","family":"Gu","sequence":"first","affiliation":[{"name":"Department of Chinese and Bilingual Studies, The Hong Kong Polytechnic University , Hong Kong 999077, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Emmanuele","family":"Chersoni","sequence":"additional","affiliation":[{"name":"Department of Chinese and Bilingual Studies, The Hong Kong Polytechnic University , Hong Kong 999077, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xing","family":"Wang","sequence":"additional","affiliation":[{"name":"Tencent AI Lab , Shenzhen 518071, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chu-Ren","family":"Huang","sequence":"additional","affiliation":[{"name":"Department of Chinese and Bilingual Studies, The Hong Kong Polytechnic University , Hong Kong 999077, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Longhua","family":"Qian","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Soochow University , Suzhou 215006, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Guodong","family":"Zhou","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Soochow University , Suzhou 215006, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2022,11,25]]},"reference":[{"key":"2022112510092267700_R1","article-title":"CORD-19: the COVID-19 Open Research Dataset","author":"Wang","year":"2020","journal-title":"ArXiv Preprint"},{"key":"2022112510092267700_R2","article-title":"Co-search: COVID-19 information retrieval with semantic search, question answering, and abstractive summarization","author":"Esteva","year":"2020","journal-title":"ArXiv Preprint"},{"key":"2022112510092267700_R3","doi-asserted-by":"crossref","first-page":"D1534","DOI":"10.1093\/nar\/gkaa952","article-title":"LitCovid: an open database of COVID-19 literature","volume":"49","author":"Chen","year":"2021","journal-title":"Nucleic Acids Research"},{"key":"2022112510092267700_R4","doi-asserted-by":"crossref","DOI":"10.1038\/d41586-020-00694-1","article-title":"Keep up with the latest coronavirus research","volume":"579","author":"Chen","year":"2020","journal-title":"Nature"},{"key":"2022112510092267700_R5","doi-asserted-by":"crossref","DOI":"10.2196\/22453","article-title":"Artificial intelligence-aided precision medicine for COVID-19: strategic areas of research and development","volume":"23","author":"Santus","year":"2021","journal-title":"Journal of Medical Internet Research"},{"key":"2022112510092267700_R6","first-page":"194","article-title":"Overview of BioASQ 2020: the eighth BioASQ challenge on large-scale biomedical semantic indexing and question answering","author":"Nentidis","year":"2020"},{"key":"2022112510092267700_R7","doi-asserted-by":"crossref","first-page":"i339","DOI":"10.1093\/bioinformatics\/btv237","article-title":"MeSHLabeler: improving the accuracy of large-scale MeSH indexing by integrating diverse evidence","volume":"31","author":"Liu","year":"2015","journal-title":"Bioinformatics"},{"key":"2022112510092267700_R8","doi-asserted-by":"crossref","DOI":"10.1093\/database\/baw042","article-title":"Chemical-induced disease relation extraction with various linguistic features","volume":"2016","author":"Gu","year":"2016","journal-title":"Database"},{"key":"2022112510092267700_R9","doi-asserted-by":"crossref","DOI":"10.1093\/database\/bax024","article-title":"Chemical-induced disease relation extraction via convolutional neural network","volume":"2017","author":"Gu","year":"2017","journal-title":"Database"},{"key":"2022112510092267700_R10","doi-asserted-by":"crossref","DOI":"10.1186\/s12859-019-2884-4","article-title":"Chemical-induced disease relation extraction via attention-based distant supervision","volume":"20","author":"Gu","year":"2019","journal-title":"BMC Bioinformatics"},{"key":"2022112510092267700_R11","first-page":"266","article-title":"Overview of the BioCreative VII LitCovid Track: multi-label topic classification for COVID-19 literature annotation","author":"Chen","year":"2021"},{"key":"2022112510092267700_R12","doi-asserted-by":"crossref","DOI":"10.1093\/database\/baac069","article-title":"Multi-label classification for biomedical literature: an overview of the BioCreative VII LitCovid Track for COVID-19 literature topic annotations","volume":"2022","author":"Chen","year":"2022","journal-title":"Database"},{"key":"2022112510092267700_R13","first-page":"326","article-title":"Team PolyU-CBSNLP at BioCreative-VII Litcovid Track: ensemble learning for COVID-19 multilabel classification","author":"Gu","year":"2021"},{"key":"2022112510092267700_R14","article-title":"Asymmetric loss for multi-label classification","author":"Ben-Baruch","year":"2020","journal-title":"ArXiv Preprint"},{"key":"2022112510092267700_R15","first-page":"268","article-title":"The NLM indexing initiative\u2019s medical text indexer","volume":"107","author":"Aronson","year":"2004","journal-title":"Medinfo"},{"key":"2022112510092267700_R16","doi-asserted-by":"crossref","first-page":"1533","DOI":"10.1093\/bioinformatics\/btz756","article-title":"FullMeSH: improving large-scale MeSH indexing with full text","volume":"36","author":"Dai","year":"2020","journal-title":"Bioinformatics"},{"key":"2022112510092267700_R17","first-page":"47","article-title":"AttentionMesh: simple, effective and interpretable automatic mesh indexer","author":"Jin","year":"2018"},{"key":"2022112510092267700_R18","doi-asserted-by":"crossref","first-page":"3794","DOI":"10.1093\/bioinformatics\/btz142","article-title":"MeSHProbeNet: a self-attentive probe net for MeSH indexing","volume":"35","author":"Xun","year":"2019","journal-title":"Bioinformatics"},{"key":"2022112510092267700_R19","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3421713","article-title":"MeSHProbeNet-P: improving large-scale MeSH indexing with personalizable MeSH probes","volume":"15","author":"Xun","year":"2020","journal-title":"ACM Transactions on Knowledge Discovery from Data"},{"key":"2022112510092267700_R20","first-page":"265","article-title":"Medical subject headings (MeSH)","volume":"88","author":"Lipscomb","year":"2000","journal-title":"Bull Med Libr Assoc."},{"key":"2022112510092267700_R21","article-title":"Overview of BioASQ 2021: the ninth BioASQ challenge on large-scale biomedical semantic indexing and question answering","author":"Anastasios","year":"2021","journal-title":"ArXiv Preprint"},{"key":"2022112510092267700_R22","first-page":"292","article-title":"Team DUT914 at BioCreative VII Litcovid Track: a BioBERT-based feature enhancement approach","author":"Tang","year":"2021"},{"key":"2022112510092267700_R23","first-page":"289","article-title":"Team DonutNLP at BioCreativeVII Litcovid Track: multi-label topic classification for COVID-19 literature annotation using the BERT-based ensemble learning approach","author":"Lin","year":"2021"},{"key":"2022112510092267700_R24","first-page":"272","article-title":"Team Bioformer at BioCreative VII LitCovid Track: multic-label topic classification for COVID-19 literature with a compact BERT model","author":"Fang","year":"2021"},{"key":"2022112510092267700_R25","doi-asserted-by":"crossref","first-page":"3388","DOI":"10.1109\/TPAMI.2020.2981890","article-title":"Imbalance problems in object detection: a review","volume":"43","author":"Kemal","year":"2021","journal-title":"IEEE Transactions on Pattern Analysis & Machine Intelligence"},{"key":"2022112510092267700_R26","doi-asserted-by":"crossref","first-page":"318","DOI":"10.1109\/TPAMI.2018.2858826","article-title":"Focal loss for dense object detection","volume":"42","author":"Lin","year":"2017","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"2022112510092267700_R27","article-title":"Ensemble learning: a survey","volume":"8","author":"Sagi","year":"2018","journal-title":"Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery"},{"key":"2022112510092267700_R28","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 Transactions on Computing for Healthcare"},{"key":"2022112510092267700_R29","doi-asserted-by":"publisher","DOI":"10.32473\/flairs.v34i1.128488","article-title":"CovidBERT-Biomedical Relation Extraction for Covid-19","volume":"34","author":"Hebbar","year":"2021","journal-title":"The International FLAIRS Conference Proceedings"},{"key":"2022112510092267700_R30","doi-asserted-by":"crossref","first-page":"1234","DOI":"10.1093\/bioinformatics\/btz682","article-title":"BioBERT: a pre-trained biomedical language representation model for biomedical text mining","volume":"36","author":"Lee","year":"2020","journal-title":"Bioinformatics"},{"key":"2022112510092267700_R31","first-page":"221","article-title":"BioM-transformers: building large biomedical language models with BERT, ALBERT and ELECTRA","author":"Alrowili","year":"2021"},{"key":"2022112510092267700_R32","first-page":"143","article-title":"BioELECTRA: pretrained biomedical text encoder using discriminators","author":"Kanakarajan","year":"2021"},{"key":"2022112510092267700_R33","doi-asserted-by":"crossref","first-page":"8342","DOI":"10.18653\/v1\/2020.acl-main.740","article-title":"Don\u2019t stop pretraining: adapt language models to domains and tasks","volume-title":"Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics","author":"Gururangan","year":"2020"},{"key":"2022112510092267700_R34","article-title":"BERT: pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018","journal-title":"ArXiv Preprint"},{"key":"2022112510092267700_R35","article-title":"Electra: pre-training text encoders as discriminators rather than generators","author":"Clark","year":"2020","journal-title":"ArXiv Preprint"},{"key":"2022112510092267700_R36","article-title":"RoBERTa: a robustly optimized BERT pretraining approach","author":"Liu","year":"2019","journal-title":"ArXiv Preprint"},{"key":"2022112510092267700_R37","first-page":"2493","article-title":"Natural language processing (almost) from scratch","volume":"12","author":"Collobert","year":"2011","journal-title":"Journal of Machine Learning Research"},{"key":"2022112510092267700_R38","article-title":"Decoupled weight decay regularization","author":"Loshchilov","year":"2017","journal-title":"ArXiv Preprint"},{"key":"2022112510092267700_R39","doi-asserted-by":"crossref","first-page":"1279","DOI":"10.1093\/jamia\/ocz085","article-title":"ML-Net: multi-label classification of biomedical texts with deep neural networks","volume":"26","author":"Du","year":"2019","journal-title":"Journal of the American Medical Informatics Association"}],"container-title":["Database"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/database\/article-pdf\/doi\/10.1093\/database\/baac103\/47263644\/baac103.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/database\/article-pdf\/doi\/10.1093\/database\/baac103\/47263644\/baac103.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,3,11]],"date-time":"2023-03-11T05:01:58Z","timestamp":1678510918000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/database\/article\/doi\/10.1093\/database\/baac103\/6846687"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,1,1]]},"references-count":39,"URL":"https:\/\/doi.org\/10.1093\/database\/baac103","relation":{},"ISSN":["1758-0463"],"issn-type":[{"type":"electronic","value":"1758-0463"}],"subject":[],"published-other":{"date-parts":[[2022,1,1]]},"published":{"date-parts":[[2022,1,1]]},"article-number":"baac103"}}