DOI QR코드

DOI QR Code

Data Sentiment Prediction through Analysis of Restaurant Consumer Reviews

맛집 소비자 리뷰 분석을 통한 데이터 감성 예측

  • Bong-Hyun Kim (Department of Computer Engineering, Seowon University)
  • 김봉현 (서원대학교 컴퓨터공학과)
  • Received : 2025.04.02
  • Accepted : 2025.05.20
  • Published : 2025.05.30

Abstract

This paper aims to analyze restaurant review data to predict consumer sentiment and support the development of targeted marketing strategies in the food service industry. Review data were collected from Google Maps and processed through morphological analysis and TF-IDF vectorization. A logistic regression model was then applied for sentiment classification. Utilizing a dataset of 149,995 reviews, the model achieved an accuracy of 87.5% in distinguishing positive and negative sentiments. GridSearchCV was used to optimize hyperparameters, and the results demonstrate the potential to understand consumer emotions and improve restaurant services and marketing strategies. The study contributes to sentiment analysis research and suggests that future work could incorporate deep learning models and multimodal data for more accurate and real-time sentiment predictions.

본 논문은 맛집 리뷰 데이터를 분석하여 소비자의 감정을 예측하고, 이를 기반으로 외식업체의 맞춤형 마케팅 전략 수립에 기여하는 것을 목적으로 한다. 연구 방법으로는 구글맵에서 리뷰 데이터를 크롤링하고, 형태소 분석과 TF-IDF 벡터화를 통해 텍스트 데이터를 수치화한 뒤, 로지스틱 회귀 모델을 적용하여 감성 분류를 수행하였다. 총 149,995개의 리뷰 데이터를 활용하였으며, 긍정과 부정 감정을 87.5%의 정확도로 예측하는 모델을 구축하였다. GridSearchCV를 통해 하이퍼파라미터를 최적화하였고, 분석 결과는 소비자 감정 이해뿐 아니라 음식점의 서비스 개선과 마케팅 전략 수립에 실질적으로 활용 가능함을 보였다. 본 연구는 실시간 감성 분석 및 딥러닝 기반 모델로의 확장을 통해 더욱 정교한 분석이 가능할 것으로 기대된다.

Keywords

References

  1. Kim, Y., & Kim, H.(2021). Opinion mining and sentiment analysis in food service reviews using deep learning. Sustainability, 13(2), 671. DOI : 10.3390/su13020671
  2. Zhang, Y., Sun, A., Rao, Y., & Liu, T.(2021). Sentiment classification with sentence-level attention mechanism in online reviews. Information Processing & Management, 58(3), 102517. DOI : 10.1016/j.ipm.2021.102517
  3. Yang, Y., Yu, J., & Chen, J.(2021). Deep learning for sentiment analysis of restaurant reviews. IEEE Access, 9, 25235–25245. DOI : 10.1109/ACCESS.2021.3057326
  4. Chatterjee, A., Agrawal, A., & Arora, A. (2021). An explainable machine learning framework for sentiment analysis of food reviews. Expert Systems with Applications, 168, 114481. DOI : 10.1016/j.eswa.2020.114481
  5. Jeon, H., Lee, D., & Kang, M.(2022). Real-time sentiment analysis for restaurant reviews using BERT and LSTM. Journal of Data and Information Science, 7(1), 40–52. DOI : 10.2478/jdis-2022-0004
  6. Cho, S., & Kim, M.(2023). Korean text sentiment classification using KoBERT. Korean Journal of Artificial Intelligence, 9(1), 21–33. DOI : 10.57151/kjai.2023.9.1.21
  7. Wang, Y., & Li, Q.(2021). A hybrid attention network for sentiment analysis of user-generated restaurant reviews. Knowledge-Based Systems, 221, 106940. DOI : 10.1016/j.knosys.2021.106940
  8. Han, J., & Lee, S.(2020). Restaurant recommendation system based on online reviews using LDA and sentiment analysis. International Journal of Computer Applications, 177(24), 12–16. DOI : 10.5120/ijca2020919785
  9. Xu, X., Liu, W., & Zhang, M.(2022). A novel review-based restaurant recommender system using transformer architecture. IEEE Transactions on Neural Networks and Learning Systems, Early Access. DOI : 10.1109/TNNLS.2022.3147891
  10. Lee, J., & Choi, H.(2023). Improving customer satisfaction through sentiment-based restaurant analysis using NLP. Asia Pacific Journal of Information Systems, 33(2), 95–110. DOI : 10.14329/apjis.2023.33.2.05
  11. Gao, S., Huang, Q., & Liu, Y. (2020). Sentiment analysis of restaurant reviews using deep learning. Journal of Physics: Conference Series, 1648(2), 022039. DOI : 10.1088/1742-6596/1648/2/022039
  12. Park, J., & Lee, S. (2021). Sentiment analysis of restaurant reviews using a convolutional neural network. Journal of Information Science Theory and Practice, 9(1), 25–39. DOI : 10.1633/JISTAP.2021.9.1.3
  13. Kim, S., & Park, H.(2022). Sentiment analysis of Korean restaurant reviews using deep learning techniques. Journal of KIISE, 49(3), 285–290. DOI : 10.5626/JOK.2022.49.3.285
  14. Lim, S., & Lee, J.(2020). Sentiment analysis of restaurant reviews using machine learning techniques. Journal of KIISE, 47(6), 567–573. DOI : 10.5626/JOK.2020.47.6.567
  15. Seo, Y., & Kim, D. (2021). Sentiment analysis of restaurant reviews using BERT. Journal of KIISE, 48(5), 512-518. DOI: 10.5626/JOK.2021.48.5.512