Abstract
Many of the texts posted on Twitter are broken sentences, and the translated sentences may not be accurate. An inaccurate translation may spoil the meaning of the original text and induce miscommunication between the poster and the reader who uses the machine translation. Since many sentences tweeted on Twitter contain emotional expressions, this study uses sentiment analysis to calculate and compare the sentiment scores of the original and translated sentences to investigate the change in sentiment before and after machine translation. As a result of using dictionaries to classify tweets before and after translation, it was found that the classification of positive sentences tended to be more likely the same before and after translation. In addition, the results of the sentiment analysis of “joy”, “like”, “relief” and “excitement” by machine learning showed that the sentiment of “joy” was particularly increased when translated from Japanese into English.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
References
Twitter. Q3 2021 Letter to Shareholders (2021)
Mohammad, S.M., Salameh, M., Kiritchenko, S.: How translation alters sentiment. J. Artif. Intell. Res. 55, 95-130 (2016)
Nakamura, A.: A Dictionary of Emotional Performance. Tokyodo Publishing, Tokyo (1993)
Plutchik, R.: The nature of emotions. Am. Sci. Res. Triangle Park 89(4), 344–350 (2001)
Yamamoto, Y., et al.: Extracting of multi-dimensional sentiment from tweets using twittism. In: Proceeding of IPSJ Kansai-Branch Convention, G-01 (2014)
Chen, B., Zhu, X.: Bilingual sentiment consistency for statistical machine translation. In: Proceedings of the 14th Conference of the European Chapter of the Association for Computational Linguistics (2014)
Saeki, M., et al.: Japanese-english target analysis of japanese adverbs focusing on emotional expression. In: Proceedings of the Association for Natural Language Processing Annual Convention, A1–9 (2005)
Tobita, H., Asada, Y.: Modern Adverbial Usage Dictionary. Tokyodo Publishing, Tokyo (1994)
Devlin, J., Chang, M.-W., Lee, K., Toutanova, K.: BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. arXiv (2018)
Gatti, L., Guerini, M., Turchi, M.: SentiWords: Deriving a high precision and high coverage lexicon for sentiment analysis. IEEE Trans. Affect. Comput. 7(4), 409–421 (2016)
Kobayashi, N., Inui, K., Matsumoto, Y., Tateishi, K., Fukushima, T.: Collecting evaluative expressions for opinion extraction. In: Su, K.-Y., Tsujii, J., Lee, J.-H., Kwong, O.Y. (eds.) IJCNLP 2004. LNCS (LNAI), vol. 3248, pp. 596–605. Springer, Heidelberg (2005). https://doi.org/10.1007/978-3-540-30211-7_63
Tago, K., Jin, Q.: Influence analysis of emotional behavious and user relationships based on twitter. Tsinghua Sci. Technol. 23(1), 104–113 (2018)
cl–tohoku/bert–base–Japanese–whole–world–masking, Hugging Face. https://huggingface.co/cl-tohoku/bert-base-japanese-whole-word-masking
Zhu, Y., et al.: Aligning Books and Movies: Towards Story-like Visual Explanations by Watching Movies and Reading Books. Arxiv (2015)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Fukuda, K., Jin, Q. (2022). Analyzing Change on Emotion Scores of Tweets Before and After Machine Translation. In: Meiselwitz, G. (eds) Social Computing and Social Media: Design, User Experience and Impact. HCII 2022. Lecture Notes in Computer Science, vol 13315. Springer, Cham. https://doi.org/10.1007/978-3-031-05061-9_21
Download citation
DOI: https://doi.org/10.1007/978-3-031-05061-9_21
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-05060-2
Online ISBN: 978-3-031-05061-9
eBook Packages: Computer ScienceComputer Science (R0)