{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,6,9]],"date-time":"2024-06-09T05:13:36Z","timestamp":1717910016043},"reference-count":21,"publisher":"MIT Press - Journals","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["TACL"],"published-print":{"date-parts":[[2018,12]]},"abstract":"<jats:p> There is often the need to perform sentiment classification in a particular domain where no labeled document is available. Although we could make use of a general-purpose off-the-shelf sentiment classifier or a pre-built one for a different domain, the effectiveness would be inferior. In this paper, we explore the possibility of building domain-specific sentiment classifiers with unlabeled documents only. Our investigation indicates that in the word embeddings learned from the unlabeled corpus of a given domain, the distributed word representations (vectors) for opposite sentiments form distinct clusters, though those clusters are not transferable across domains. Exploiting such a clustering structure, we are able to utilize machine learning algorithms to induce a quality domain-specific sentiment lexicon from just a few typical sentiment words (\u201cseeds\u201d). An important finding is that simple linear model based supervised learning algorithms (such as linear SVM) can actually work better than more sophisticated semi-supervised\/transductive learning algorithms which represent the state-of-the-art technique for sentiment lexicon induction. The induced lexicon could be applied directly in a lexicon-based method for sentiment classification, but a higher performance could be achieved through a two-phase bootstrapping method which uses the induced lexicon to assign positive\/negative sentiment scores to unlabeled documents first, a nd t hen u ses those documents found to have clear sentiment signals as pseudo-labeled examples to train a document sentiment classifier v ia supervised learning algorithms (such as LSTM). On several benchmark datasets for document sentiment classification, our end-to-end pipelined approach which is overall unsupervised (except for a tiny set of seed words) outperforms existing unsupervised approaches and achieves an accuracy comparable to that of fully supervised approaches. <\/jats:p>","DOI":"10.1162\/tacl_a_00020","type":"journal-article","created":{"date-parts":[[2018,12,10]],"date-time":"2018-12-10T19:32:50Z","timestamp":1544470370000},"page":"269-285","source":"Crossref","is-referenced-by-count":7,"title":["Bootstrap Domain-Specific Sentiment Classifiers from Unlabeled Corpora"],"prefix":"10.1162","volume":"6","author":[{"given":"Andrius","family":"Mudinas","sequence":"first","affiliation":[{"name":"Department of Computer Science and Information Systems, Birkbeck, University of London, London WC1E 7HX, UK,"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dell","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Information Systems, Birkbeck, University of London, London WC1E 7HX, UK,"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mark","family":"Levene","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Information Systems, Birkbeck, University of London, London WC1E 7HX, UK,"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"281","reference":[{"issue":"6","key":"p_1","doi-asserted-by":"crossref","first-page":"802","DOI":"10.1002\/asi.20553","volume":"58","author":"Argamon Shlomo","year":"2007","journal-title":"Journal of the American Society for Information Science and Technology (JASIST)"},{"issue":"8","key":"p_4","doi-asserted-by":"crossref","first-page":"1719","DOI":"10.1109\/TKDE.2012.103","volume":"25","author":"Bollegala Danushka","year":"2013","journal-title":"IEEE Transactions on Knowledge and Data Engineering (TKDE)"},{"issue":"1","key":"p_5","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.jocs.2010.12.007","volume":"2","author":"Bollen Johan","year":"2011","journal-title":"Journal of Computational Science"},{"issue":"1","key":"p_6","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1175\/1520-0493(1950)078<0001:VOFEIT>2.0.CO;2","volume":"78","author":"Brier Glenn W.","year":"1950","journal-title":"Monthly Weather Review"},{"issue":"1","key":"p_16","doi-asserted-by":"crossref","first-page":"1","DOI":"10.2200\/S00762ED1V01Y201703HLT037","volume":"10","author":"Goldberg Yoav","year":"2017","journal-title":"Synthesis Lectures on Human Language Technologies"},{"issue":"10","key":"p_17","doi-asserted-by":"crossref","first-page":"2222","DOI":"10.1109\/TNNLS.2016.2582924","volume":"28","author":"Greff Klaus","year":"2017","journal-title":"IEEE Transactions on Neural Networks and Learning Systems (TNNLS)"},{"key":"p_20","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"issue":"2","key":"p_22","doi-asserted-by":"crossref","first-page":"415","DOI":"10.1109\/72.991427","volume":"13","author":"Hsu Chih-Wei","year":"2002","journal-title":"IEEE Transactions on Neural Networks (TNN)"},{"issue":"2","key":"p_29","doi-asserted-by":"crossref","first-page":"100","DOI":"10.1111\/j.1467-8640.2006.00276.x","volume":"22","author":"Koppel Moshe","year":"2006","journal-title":"Computational Intelligence"},{"issue":"7553","key":"p_30","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1038\/nature14539","volume":"521","author":"LeCun Yann","year":"2015","journal-title":"Nature"},{"issue":"1","key":"p_33","doi-asserted-by":"crossref","first-page":"1","DOI":"10.2200\/S00416ED1V01Y201204HLT016","volume":"5","author":"Liu Bing","year":"2012","journal-title":"Synthesis Lectures on Human Language Technologies"},{"issue":"3","key":"p_35","doi-asserted-by":"crossref","first-page":"462","DOI":"10.1109\/TPAMI.2015.2452921","volume":"38","author":"Loog Marco","year":"2016","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)"},{"issue":"1","key":"p_40","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1080\/01690969108406936","volume":"6","author":"Miller George A","year":"1991","journal-title":"Language and Cognitive Processes"},{"key":"p_51","doi-asserted-by":"publisher","DOI":"10.1162\/coli_a_00034"},{"issue":"1","key":"p_53","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1140\/epjds\/s13688-015-0062-0","volume":"5","author":"Ribeiro Filipe N","year":"2016","journal-title":"EPJ Data Science"},{"issue":"1","key":"p_57","first-page":"1929","volume":"15","author":"Srivastava Nitish","year":"2014","journal-title":"Journal of Machine Learning Research (JMLR)"},{"issue":"12","key":"p_60","doi-asserted-by":"crossref","first-page":"2544","DOI":"10.1002\/asi.21416","volume":"61","author":"Thelwall Mike","year":"2010","journal-title":"Journal of the American Society for Information Science and Technology (JASIST)"},{"issue":"1","key":"p_61","doi-asserted-by":"crossref","first-page":"163","DOI":"10.1002\/asi.21662","volume":"63","author":"Thelwall Mike","year":"2012","journal-title":"Journal of the American Society for Information Science and Technology (JASIST)"},{"key":"p_63","first-page":"2579","volume":"9","author":"van der Maaten Laurens","year":"2008","journal-title":"Journal of Machine Learning Research (JMLR)"},{"issue":"4","key":"p_65","doi-asserted-by":"crossref","first-page":"1191","DOI":"10.3758\/s13428-012-0314-x","volume":"45","author":"Warriner Amy Beth","year":"2013","journal-title":"Behavior Research Methods"},{"issue":"3","key":"p_66","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1109\/MIS.2013.27","volume":"28","author":"Xia Rui","year":"2013","journal-title":"IEEE Intelligent Systems"}],"container-title":["Transactions of the Association for Computational Linguistics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mitpressjournals.org\/doi\/pdf\/10.1162\/tacl_a_00020","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,3,12]],"date-time":"2021-03-12T21:37:55Z","timestamp":1615585075000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/tacl\/article\/43430"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2018,12]]},"references-count":21,"alternative-id":["10.1162\/tacl_a_00020"],"URL":"https:\/\/doi.org\/10.1162\/tacl_a_00020","relation":{},"ISSN":["2307-387X"],"issn-type":[{"value":"2307-387X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2018,12]]}}}