Abstract
In recent years, customer reviews have been attracting attention as data representing real consumer feedback. On the other hand, it is said that it is difficult to identify the detailed contents from the text. In this study, we aim to understand the detailed contents of customer reviews from the data of customer reviews of accommodation facilities. Specifically, we collected tweets using keywords related to accommodation reviews as search queries, and used them to create a sentence classification model by fine-tuning a previously developed Japanese version of BERT model. Sentence classification was conducted by applying this model to the customer review data, which was separated into sentences. In addition, sentiment analysis was conducted to identify whether the sentences were positive or negative.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Booking.com official site. https://www.booking.com/index.ja.html (2 Aug 2023 author checked)
Expedia official site. https://www.expedia.co.jp/ (2 Aug 2023 author checked)
Rakuten Travel official site. https://travel.rakuten.co.jp/ (2 Aug 2023 author checked)
Jalan net official site. https://www.jalan.net/ (2 Aug 2023 author checked)
JTB Tourism Research & Consulting Co. Survey on Smartphone Use and Travel Consumption (2019)
Yamashita, K., Kumoi, G., Hasumoto, K.: Masayuki Goto a study of analytical model of product review data based on features extracted by BERT”. Inf. Process. Soc. Japan 1, 841–842 (2022)
Aslam, N., Ramay, W.Y., Xia, K., Sarwar, N.: Convolutional neural network-based classification of app reviews. IEEE Access 8, 185619–185628 (2020)
Classification Model for Application Review Training with Forums,” The Institute of Electronics, Information and Communication Engineers. D, pp 669–678 (2022)
Devlin, J., Chang, M., Lee, K., Toutanova K.: BERT: Pretraining of deep bidirectional transformers for language understanding. In: Proceedings of Conference North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp. 4171–4186 (2019)
Rakuten Group, Inc. Rakuten Dataset. Informatics Research Data Repository, National Institute of Informatics (2014). https://doi.org/10.32130/idr.2.0 (2 Aug 2023 author checked)
Pretrained Japanese BERT models released | Tohoku NLP Lab https://www.nlp.ecei.tohoku.ac.jp/news-release/3284/(2023/2/08 author checked)
Hugging Face:https://huggingface.co/daigo/bert-base-japanese-sentiment (2 Aug 2023 author checked)
Forbes Travel Guide Star Rated Hotels, Restaurants & Spas https://www.forbestravelguide.com/ (2 Aug 2023 author checked)
Acknowledgment
In this paper, we used “Rakuten Dataset”(https://rit.rakuten.com/data_release/) provided by Rakuten Group, Inc. Via IDR Dataset Service of National Institute of Informatics. This work was supported by JSPS KAKENHI Grant Number 21H04600 and 21K13385.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Iwanade, E., Namatame, T., Otake, K. (2023). Identification of Evaluation Items in Consumer Reviews Using Natural Language Processing Models with Social Media Information. In: Coman, A., Vasilache, S. (eds) Social Computing and Social Media. HCII 2023. Lecture Notes in Computer Science, vol 14025. Springer, Cham. https://doi.org/10.1007/978-3-031-35915-6_38
Download citation
DOI: https://doi.org/10.1007/978-3-031-35915-6_38
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-35914-9
Online ISBN: 978-3-031-35915-6
eBook Packages: Computer ScienceComputer Science (R0)