{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,20]],"date-time":"2026-03-20T21:29:45Z","timestamp":1774042185183,"version":"3.50.1"},"reference-count":35,"publisher":"Emerald","issue":"3","license":[{"start":{"date-parts":[[2020,9,21]],"date-time":"2020-09-21T00:00:00Z","timestamp":1600646400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.emerald.com\/insight\/site-policies"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IJWIS"],"published-print":{"date-parts":[[2020,9,21]]},"abstract":"<jats:sec><jats:title content-type=\"abstract-subheading\">Purpose<\/jats:title><jats:p>Recently, Unicode has been standardized with the penetration of social networking services, the use of emojis has become common. Emojis, as they are also known, are most effective in expressing emotions in sentences. Sentiment analysis in natural language processing manually labels emotions for sentences. The authors can predict sentiment using emoji of text posted on social media without labeling manually. The purpose of this paper is to propose a new model that learns from sentences using emojis as labels, collecting English and Japanese tweets from Twitter as the corpus. The authors verify and compare multiple models based on attention long short-term memory (LSTM) and convolutional neural networks (CNN) and Bidirectional Encoder Representations from Transformers (BERT).<\/jats:p><\/jats:sec><jats:sec><jats:title content-type=\"abstract-subheading\">Design\/methodology\/approach<\/jats:title><jats:p>The authors collected 2,661 kinds of emoji registered as Unicode characters from tweets using Twitter application programming interface. It is a total of 6,149,410 tweets in Japanese. First, the authors visualized a vector space produced by the emojis by Word2Vec. In addition, the authors found that emojis and similar meaning words of emojis are adjacent and verify that emoji can be used for sentiment analysis. Second, it involves entering a line of tweets containing emojis, learning and testing with that emoji as a label. The authors compared the BERT model with the conventional models [CNN, FastText and Attention bidirectional long short-term memory (BiLSTM)] that were high scores in the previous study.<\/jats:p><\/jats:sec><jats:sec><jats:title content-type=\"abstract-subheading\">Findings<\/jats:title><jats:p>Visualized the vector space of Word2Vec, the authors found that emojis and similar meaning words of emojis are adjacent and verify that emoji can be used for sentiment analysis. The authors obtained a higher score with BERT models compared to the conventional model. Therefore, the sophisticated experiments demonstrate that they improved the score over the conventional model in two languages. General emoji prediction is greatly influenced by context. In addition, the score may be lowered due to a misunderstanding of meaning. By using BERT based on a bi-directional transformer, the authors can consider the context.<\/jats:p><\/jats:sec><jats:sec><jats:title content-type=\"abstract-subheading\">Practical implications<\/jats:title><jats:p>The authors can find emoji in the output words by typing a word using an input method editor (IME). The current IME only considers the most latest inputted word, although it is possible to recommend emojis considering the context of the inputted sentence in this study. Therefore, the research can be used to improve IME performance in the future.<\/jats:p><\/jats:sec><jats:sec><jats:title content-type=\"abstract-subheading\">Originality\/value<\/jats:title><jats:p>In the paper, the authors focus on multilingual emoji prediction. This is the first attempt of comparison at emoji prediction between Japanese and English. In addition, it is also the first attempt to use the BERT model based on the transformer for predicting limited emojis although the transformer is known to be effective for various NLP tasks. The authors found that a bidirectional transformer is suitable for emoji prediction.<\/jats:p><\/jats:sec>","DOI":"10.1108\/ijwis-09-2019-0042","type":"journal-article","created":{"date-parts":[[2020,9,24]],"date-time":"2020-09-24T06:50:39Z","timestamp":1600930239000},"page":"265-280","source":"Crossref","is-referenced-by-count":31,"title":["Multilingual emoji prediction using BERT for sentiment analysis"],"prefix":"10.1108","volume":"16","author":[{"given":"Toshiki","family":"Tomihira","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Atsushi","family":"Otsuka","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Akihiro","family":"Yamashita","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tetsuji","family":"Satoh","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"140","reference":[{"key":"key2020100707570801900_ref001","doi-asserted-by":"crossref","first-page":"405","DOI":"10.18653\/v1\/S18-1062","article-title":"The dabblers at SemEval-2018 task 2: multilingual emoji prediction","volume-title":"Proceedings of The 12th International Workshop on Semantic Evaluation","year":"2018"},{"key":"key2020100707570801900_ref002","first-page":"105","article-title":"Are emojis predictable?","year":"2017"},{"key":"key2020100707570801900_ref003","article-title":"Multi-task emoji learning","year":"2018"},{"key":"key2020100707570801900_ref004","first-page":"679","article-title":"Multimodal emoji prediction","year":"2018"},{"key":"key2020100707570801900_ref005","doi-asserted-by":"crossref","first-page":"4766","DOI":"10.18653\/v1\/D18-1508","article-title":"Interpretable emoji prediction via label-wise attention LSTMs","volume-title":"Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing","year":"2018"},{"key":"key2020100707570801900_ref006","doi-asserted-by":"crossref","first-page":"24","DOI":"10.18653\/v1\/S18-1003","article-title":"SemEval 2018 task 2: multilingual emoji prediction","volume-title":"Proceedings of The 12th International Workshop on Semantic Evaluation","year":"2018"},{"key":"key2020100707570801900_ref007","doi-asserted-by":"crossref","first-page":"438","DOI":"10.18653\/v1\/S18-1069","article-title":"NTUA-SLP at SemEval-2018 task 2: predicting emojis using RNNs with context-aware attention","volume-title":"Proceedings of The 12th International Workshop on Semantic Evaluation","year":"2018"},{"key":"key2020100707570801900_ref008","doi-asserted-by":"crossref","first-page":"245","DOI":"10.18653\/v1\/S18-1037","article-title":"NTUA-SLP at SemEval-2018 task 1: predicting affective content in tweets with deep attentive RNNs and transfer learning","volume-title":"Proceedings of The 12th International Workshop on Semantic Evaluation","year":"2018"},{"key":"key2020100707570801900_ref009","doi-asserted-by":"crossref","first-page":"135","DOI":"10.1162\/tacl_a_00051","article-title":"Enriching word vectors with subword information","volume":"5","year":"2017","journal-title":"Transactions of the Association for Computational Linguistics"},{"key":"key2020100707570801900_ref010","doi-asserted-by":"crossref","first-page":"251","DOI":"10.1145\/3308558.3313600","article-title":"Emoji-powered representation learning for cross-lingual sentiment classification","volume-title":"The World Wide Web Conference","year":"2019"},{"key":"key2020100707570801900_ref011","article-title":"Contrastive learning of emoji-based representations for resource-poor languages","year":"2018"},{"key":"key2020100707570801900_ref012","first-page":"4171","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","year":"2019"},{"key":"key2020100707570801900_ref013","first-page":"48","article-title":"emoji2vec: learning emoji representations from their description","year":"2016"},{"key":"key2020100707570801900_ref014","first-page":"1615","article-title":"Using millions of emoji occurrences to learn any-domain representations for detecting sentiment, emotion and sarcasm","volume-title":"Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing","year":"2017"},{"key":"key2020100707570801900_ref015","article-title":"Learning to forget: Continual prediction with lstm","year":"1999"},{"key":"key2020100707570801900_ref016","doi-asserted-by":"crossref","first-page":"167","DOI":"10.1016\/j.procs.2018.05.200","article-title":"Opinion mining on emojis using deep learning techniques","volume":"132","year":"2018","journal-title":"Procedia Computer Science"},{"key":"key2020100707570801900_ref017","doi-asserted-by":"publisher","article-title":"Convolutional neural networks for sentence classification","year":"2014","DOI":"10.3115\/v1\/D14-1181"},{"key":"key2020100707570801900_ref018","doi-asserted-by":"crossref","first-page":"497","DOI":"10.18653\/v1\/S18-1080","article-title":"Tweety at SemEval-2018 task 2: predicting emojis using hierarchical attention neural networks and support vector machine","volume-title":"Proceedings of The 12th International Workshop on Semantic Evaluation","year":"2018"},{"key":"key2020100707570801900_ref019","first-page":"1097","article-title":"Imagenet classification with deep convolutional neural networks","volume":"1","year":"2012","journal-title":"Advances in Neural Information Processing Systems"},{"key":"key2020100707570801900_ref020","unstructured":"Kudo, T. (2009), \u201cMecab yetanother part-of-speed and morphological analyzer\u201d, available at: http:\/\/mecab.sourceforge.net\/"},{"key":"key2020100707570801900_ref021","doi-asserted-by":"crossref","first-page":"433","DOI":"10.18653\/v1\/S18-1068","article-title":"ECNU at SemEval-2018 task 2: leverage traditional NLP features and neural networks methods to address twitter emoji prediction task","volume-title":"Proceedings of The 12th International Workshop on Semantic Evaluation","year":"2018"},{"key":"key2020100707570801900_ref022","article-title":"Efficient estimation of word representations in vector space","year":"2013"},{"key":"key2020100707570801900_ref023","doi-asserted-by":"crossref","first-page":"305","DOI":"10.18653\/v1\/S18-1045","article-title":"DeepMiner at SemEval-2018 task 1: emotion intensity recognition using deep representation learning","volume-title":"Proceedings of The 12th International Workshop on Semantic Evaluation","year":"2018"},{"key":"key2020100707570801900_ref024","doi-asserted-by":"crossref","first-page":"628","DOI":"10.18653\/v1\/S18-1102","article-title":"Binarizer at SemEval-2018 task 3: parsing dependency and deep learning for irony detection","volume-title":"Proceedings of The 12th International Workshop on Semantic Evaluation","year":"2018"},{"key":"key2020100707570801900_ref025","doi-asserted-by":"crossref","first-page":"264","DOI":"10.18653\/v1\/S18-1039","article-title":"PlusEmo2Vec at SemEval-2018 task 1: exploiting emotion knowledge from emoji and #hashtags","volume-title":"Proceedings of The 12th International Workshop on Semantic Evaluation","year":"2018"},{"key":"key2020100707570801900_ref026","first-page":"205","article-title":"Improvement of Japanese parsing accuracy with bert","volume-title":"The Association for Natural Language Processing 25th annual conference","year":"2019"},{"key":"key2020100707570801900_ref027","first-page":"380","article-title":"Utilizing BERT for aspect-based sentiment analysis via constructing auxiliary sentence","year":"2019"},{"key":"key2020100707570801900_ref028","article-title":"Lstm neural networks for language modeling","volume-title":"Thirteenth annual conference of the international speech communication association","year":"2012"},{"key":"key2020100707570801900_ref029","first-page":"289","article-title":"What does your tweet emotion mean?: neural emoji prediction for sentiment analysis","volume-title":"Proceedings of the 20th International Conference on Information Integration and Web-based Applications and Services","year":"2018"},{"key":"key2020100707570801900_ref030","first-page":"2579","article-title":"Visualizing data using t-sne","volume":"9","year":"2008","journal-title":"Journal of Machine Learning Research"},{"key":"key2020100707570801900_ref031","first-page":"5998","article-title":"Attention is all you need","volume-title":"Advances in Neural Information Processing Systems","year":"2017"},{"key":"key2020100707570801900_ref032","first-page":"132","article-title":"Predict emoji combination with retrieval strategy","volume-title":"The Association for Natural Language Processing 25th annual conference","year":"2019"},{"key":"key2020100707570801900_ref033","doi-asserted-by":"crossref","first-page":"551","DOI":"10.18653\/v1\/S19-2099","article-title":"BNU-HKBU UIC NLP team 2 at SemEval-2019 task 6: detecting offensive language using BERT model","volume-title":"Proceedings of the 13th International Workshop on Semantic Evaluation","year":"2019"},{"key":"key2020100707570801900_ref034","first-page":"253","article-title":"A sensitivity analysis of (and practitioners\u2019 guide to) convolutional neural networks for sentence classification","year":"2017"},{"key":"key2020100707570801900_ref035","doi-asserted-by":"crossref","first-page":"788","DOI":"10.18653\/v1\/S19-2138","article-title":"UM-IU@LING at SemEval-2019 task 6: identifying offensive tweets using BERT and SVMs","volume-title":"Proceedings of the 13th International Workshop on Semantic Evaluation","year":"2019"}],"container-title":["International Journal of Web Information Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.emerald.com\/insight\/content\/doi\/10.1108\/IJWIS-09-2019-0042\/full\/xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.emerald.com\/insight\/content\/doi\/10.1108\/IJWIS-09-2019-0042\/full\/html","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,24]],"date-time":"2025-07-24T22:24:16Z","timestamp":1753395856000},"score":1,"resource":{"primary":{"URL":"http:\/\/www.emerald.com\/ijwis\/article\/16\/3\/265-280\/165670"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,9,21]]},"references-count":35,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2020,9,21]]}},"alternative-id":["10.1108\/IJWIS-09-2019-0042"],"URL":"https:\/\/doi.org\/10.1108\/ijwis-09-2019-0042","relation":{},"ISSN":["1744-0084","1744-0084"],"issn-type":[{"value":"1744-0084","type":"print"},{"value":"1744-0084","type":"print"}],"subject":[],"published":{"date-parts":[[2020,9,21]]}}}