Information Systems Review (경영정보학연구)

The Korea Society of Management Information Systems (한국경영정보학회)
권호 표지
DOI QR코드

DOI QR Code

A Study of Deep Learning-based Personalized Recommendation Service for Solving Online Hotel Review and Rating Mismatch Problem

온라인 호텔 리뷰와 평점 불일치 문제 해결을 위한 딥러닝 기반 개인화 추천 서비스 연구

  • Qinglong Li (Department of Big Data Analytics, Graduate School, Kyung Hee University) ;
  • Shibo Cui (Department of Big Data Analytics, Graduate School, Kyung Hee University) ;
  • Byunggyu Shin (Department of Business Administration, Graduate School, Kyung Hee University) ;
  • Jaekyeong Kim (School of Management & Department of Big Data Analytics, Kyung Hee University)
  • 이청용 (경희대학교 대학원 빅데이터응용학과) ;
  • 최사박 (경희대학교 대학원 빅데이터응용학과) ;
  • 신병규 (경희대학교 대학원 경영학과) ;
  • 김재경 (경희대학교 경영대학/빅데이터응용학과)
  • Received : 2021.03.11
  • Accepted : 2021.05.31
  • Published : 2021.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

Global e-commerce websites offer personalized recommendation services to gain sustainable competitiveness. Existing studies have offered personalized recommendation services using quantitative preferences such as ratings. However, offering personalized recommendation services using only quantitative data has raised the problem of decreasing recommendation performance. For example, a user gave a five-star rating but wrote a review that the user was unsatisfied with hotel service and cleanliness. In such cases, has problems where quantitative and qualitative preferences are inconsistent. Recently, a growing number of studies have considered review data simultaneously to improve the limitations of existing personalized recommendation service studies. Therefore, in this study, we identify review and rating mismatches and build a new user profile to offer personalized recommendation services. To this end, we use deep learning algorithms such as CNN, LSTM, CNN + LSTM, which have been widely used in sentiment analysis studies. And extract sentiment features from reviews and compare with quantitative preferences. To evaluate the performance of the proposed methodology in this study, we collect user preference information using real-world hotel data from the world's largest travel platform TripAdvisor. Experiments show that the proposed methodology in this study outperforms the existing other methodologies, using only existing quantitative preferences.

세계적인 전자상거래 기업들은 지속 가능한 경쟁력을 확보하기 위해 사용자 맞춤형 추천 서비스를 제공하고 있다. 기존 관련 연구에서는 주로 평점, 구매 여부 등 정량적 선호도 정보를 사용하여 개인화 추천 서비스를 제공하였다. 하지만 이와 같은 정량적 선호도 정보를 사용하여 개인화 추천 서비스를 제공하면 추천 성능이 저하될 수 있다는 문제점이 제기되고 있다. 호텔을 이용한 사용자가 호텔 서비스, 청결 상태 등에 대하여 만족하지 못한다고 리뷰를 작성하였으나 선호도 평점 5점을 부여했을 때 정량적 선호도(평점)와 정성적 선호도(리뷰)가 불일치한 문제가 발생할 수 있다. 따라서 본 연구에서는 정량적 선호도 정보와 정성적 선호도 정보가 일치하는지를 확인하고 이를 바탕으로 선호도 정보가 일치하는 사용자를 바탕으로 새로운 프로파일을 구축하여 개인화 추천 서비스를 제공하고자 한다. 리뷰에서 정성적 선호도를 추출하기 위해 자연어 처리 관련 연구에서 널리 사용되고 있는 CNN, LSTM, CNN + LSTM 등 딥러닝 기법을 사용하여 감성분석 모델을 구축하였다. 이를 통해 사용자가 작성한 리뷰에서 정성적 선호도 정보를 정교하게 추출하여 정량적 선호도 정보와 비교하였다. 본 연구에서 제안한 추천 방법론의 성능을 평가하기 위해 세계 최대 여행 플랫폼 TripAdvisor에서 실제 호텔을 이용한 사용자 선호도 정보를 수집하여 사용하였다. 실험 결과 본 연구에서 제안한 추천 방법론이 기존의 정량적 선호도만을 고려하는 추천 방법론보다 우수한 추천 성능을 나타냄을 확인할 수 있었다.

Keywords

References

  1. 박승택, 성인재, 서상원, 황지수, 노지성, 김대원, "기계학습 기반의 뉴스 추천 서비스 구조와 그 효과에 대한 고찰: 카카오의 루빅스를 중심으로", 사이버커뮤니케이션학보, 제34권, 제1호, 2017, pp. 5-48.
  2. 배상현, 최병구, "변이형 오토인코더와 어텐션메커니즘을 결합한 차트기반 주가 예측", Information Systems Review, 제23권, 제1호, 2021, pp. 23-43. https://doi.org/10.14329/isr.2021.23.1.023
  3. 손지은, 김성범, 김현중, 조성준, "추천 시스템 기법 연구동향 분석", 대한산업공학회지, 제41권, 제2호, 2015, pp. 185-208. https://doi.org/10.7232/JKIIE.2015.41.2.185
  4. 전병국, 안현철, "사용자 리뷰 마이닝을 결합한 협업 필터링 시스템: 스마트폰 앱 추천에의 응용", 지능정보연구, 제21권, 제2호, 2015, pp. 1-18. https://doi.org/10.13088/JIIS.2015.21.2.01
  5. 현지연, 유상이, 이상용, "평점과 리뷰 텍스트 감성분석을 결합한 추천시스템 향상 방안 연구", 지능정보연구, 제25권, 제1권, 2019, pp. 219-239. https://doi.org/10.13088/JIIS.2019.25.1.219
  6. 조승연, 최지은, 이규현, 김희웅, "고객 온라인 구매후기를 활용한 추천시스템 개발 및 적용", Information Systems Review, 제17권, 제3호, 2015, pp. 95-111. https://doi.org/10.14329/isr.2015.17.3.095
  7. Adomavicius, G. and A. Tuzhilin, "Personalization technologies: A process-oriented perspective", Communications of the ACM, Vol.48, No.10, 2005, pp. 83-90. https://doi.org/10.1145/1089107.1089109
  8. Alhuzali, H., M. Elaraby, and M. Abdul-Mageed, "Ubc-nlp at iest 2018: Learning implicit emotion with an ensemble of language models", Proceedings of the 9th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis, 2018, pp. 342-347.
  9. Annett, M. and G. Kondrak, "A comparison of sentiment analysis techniques: Polarizing movie blogs", Conference of the Canadian Society for Computational Studies of Intelligence, 2008, pp. 25-35.
  10. Bennett, J. and S. Lanning, "The netflix prize", Proceedings of KDD Cup and Workshop, 2007, p. 35.
  11. Borgonovo, E. and E. Plischke, "Sensitivity analysis: A review of recent advances", European Journal of Operational Research, Vol.2248, No.3, 2016, pp. 869-887. https://doi.org/10.1016/j.ejor.2015.06.032
  12. Cakir, B. and E. Dogdu, "Malware classification using deep learning methods", Proceedings of the ACMSE 2018 Conference, 2018, pp. 1-5.
  13. Cheng, Z., Y. Ding, L. Zhu, and M. Kankanhalli, "Aspect-aware latent factor model: Rating prediction with ratings and reviews", Proceedings of the 2018 World Wide Web Conference, 2018, pp. 639-648.
  14. Choi, I. Y., M. G. Oh, J. K. Kim, and Y. U. Ryu, "Collaborative filtering with facial expressions for online video recommendation", International Journal of Information Management, Vol.36, No.3, 2016, pp. 397-402. https://doi.org/10.1016/j.ijinfomgt.2016.01.005
  15. Christopher Frey, H. and S. R. Patil, "Identification and review of sensitivity analysis methods", Risk Analysis, Vol.22, No.3, 2002, pp. 553-578. https://doi.org/10.1111/0272-4332.00039
  16. Das, A. S., M. Datar, A. Garg, and S. Rajaram, "Google news personalization: scalable online collaborative filtering", Proceedings of the 16th International Conference on World Wide Web, 2007, pp. 271-280.
  17. Faker, O. and E. Dogdu, "Intrusion detection using big data and deep learning techniques", Proceedings of the 2019 ACM Southeast Conference, 2019, pp. 86-93.
  18. Ferguson, P., N. O'Hare, M. Davy, A. Bermingham, P. Sheridan, C. Gurrin, and F. meaton, "Exploring the use of paragraph-level annotations for sentiment analysis of financial blogs", Workshop on Opinion Mining and Sentiment Analysis, 2009.
  19. Garcia-Cumbreras, M. A., A. Montejo-Raez, and M. C. Diaz-Galiano, "Pessimists and optimists: Improving collaborative filtering through sentiment analysis", Expert Systems With Applications, Vol.40, No.17, 2013, pp. 6758-6765. https://doi.org/10.1016/j.eswa.2013.06.049
  20. Hassan, A. and A. Mahmood, "Deep learning approach for sentiment analysis of short texts", 2017 3rd International Conference on Control, Automation and Robotics (ICCAR), 2017, pp. 705-710.
  21. Herlocker, J. L., J. A. Konstan, and J. Riedl, "Explaining collaborative filtering recommendations", Proceedings of the 2000 ACM Conference on Computer Supported Cooperative Work, 2000, pp. 241-250.
  22. Hofmann, T., "Latent semantic models for collaborative filtering", ACM Transactions on Information Systems (TOIS), Vol.22, No.1, 2004, pp. 89-115. https://doi.org/10.1145/963770.963774
  23. Jangid, H., S. Singhal, R. R. Shah, and R. Zimmermann, "Aspect-based financial sentiment analysis using deep learning", Companion Proceedings of the The Web Conference 2018, 2018, pp. 1961-1966.
  24. Jakob, N., S. H. Weber, M. C. Muller, and I. Gurevych, "Beyond the stars: exploiting free-text user reviews to improve the accuracy of movie recommendations", Proceedings of the 1st International CIKM Workshop on Topic-sentiment Analysis for Mass Opinion, 2009, pp. 57-64
  25. Kazmaier, J. and J. H. van Vuuren, "A generic framework for sentiment analysis: Leveraging opinion-bearing data to inform decision making", Decision Support Systems, Vol.135, 2020, p. 113304.
  26. Kennedy, A. and D. Inkpen, "Sentiment classification of movie reviews using contextual valence shifters", Computational Intelligence, Vol.22, No.2, 2006, pp. 110-125. https://doi.org/10.1111/j.1467-8640.2006.00277.x
  27. Kim, H. K., J. K. Kim, and Y. U. Ryu, "Personalized recommendation over a customer network for ubiquitous shopping", IEEE Transactions on Services Computing, Vol.2, No.2, 2009, pp. 140-151. https://doi.org/10.1109/TSC.2009.7
  28. Kim, J. K., H. K. Kim, H. Y. Oh, and Y. U. Ryu, "A group recommendation system for online communities", International Journal of Information Management, Vol.30, No.3, 2010, pp. 212-219. https://doi.org/10.1016/j.ijinfomgt.2009.09.006
  29. Kim, Y., "Convolutional Neural Networks for Sentence Classification", Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2014, pp. 1746-1751.
  30. Kingma, D. P. and J. Ba, "Adam: A method for stochastic optimization", Proceedings of the International Conference on Learning Representations (ICLR), 2015.
  31. Konig, A. C. and E. Brill, "Reducing the human overhead in text categorization", Proceedings of the 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2006, pp. 598-603.
  32. Konstan, J. A., B. N. Miller, D. Maltz, J. L. Herlocker, L. R. Gordon, and J. Riedl, "Grouplens: Applying collaborative filtering to usenet news", Communications of the ACM, Vol.40, No.3, 1997, pp. 77-87. https://doi.org/10.1145/245108.245126
  33. Lai, S., L. Xu, K. Liu, and J. Zhao, "Recurrent convolutional neural networks for text classification", Twenty-Ninth AAAI Conference on Artificial Intelligence, 2015, pp. 2267-2273
  34. Lei, X., X. Qian, and G. Zhao, "Rating prediction based on social sentiment from textual reviews", IEEE Transactions on Multimedia, Vol.18, No.9, 2016, pp. 1910-1921. https://doi.org/10.1109/TMM.2016.2575738
  35. Leung, C. W., S. C. Chan, and F. L. Chung, "Integrating collaborative filtering and sentiment analysis: A rating inference approach", Proceedings of the ECAI 2006 Workshop on Recommender Systems, 2006, pp. 62-66
  36. Levi, A., O. Mokryn, C. Diot, and N. Taft, "Finding a needle in a haystack of reviews: cold start context-based hotel recommender system", Proceedings of the Sixth ACM Conference on Recommender Systems, 2012, pp. 115-122
  37. Li, X., M. Wang, and T. P. Liang, "A multi-theoretical kernel-based approach to social network-based recommendation", Decision Support Systems, Vol.65, 2014, pp. 95-104 https://doi.org/10.1016/j.dss.2014.05.006
  38. Linden, G., B. Smith, and J. York, "Amazon. com recommendations: Item-to-item collaborative filtering", IEEE Internet Computing, Vol.7, No.1, 2003, pp. 76-80. https://doi.org/10.1109/MIC.2003.1167344
  39. Moon, H. S., J. H. Yoon, I. Y. Choi, and J. K. Kim, "An Exploratory Study of Collaborative Filtering Techniques to Analyze the Effect of Information Amount", Asia Pacific Journal of Information Systems, Vol.27, No.2, 2017, pp. 126-138. https://doi.org/10.14329/apjis.2017.27.2.126
  40. Moshfeghi, Y., B. Piwowarski, and J. M. Jose, "Handling data sparsity in collaborative filtering using emotion and semantic based features", Proceedings of the 34th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2011, pp. 625-634.
  41. Nguyen, T. T., P. M. Hui, F. M. Harper, L. Terveen, and J. A. Konstan, "Exploring the filter bubble: the effect of using recommender systems on content diversity", Proceedings of the 23rd International Conference on World Wide Web, 2014, pp. 677-686.
  42. Ombabi, A. H., W. Ouarda, and A. M. Alimi, "Deep learning CNN-LSTM framework for Arabic sentiment analysis using textual information shared in social networks", Social Network Analysis and Mining, Vol.10, No.1, 2020, pp. 1-13. https://doi.org/10.1007/s13278-020-00668-1
  43. Pak, A. and P. Paroubek, "Twitter as a corpus for sentiment analysis and opinion mining", Proceedings of the Seventh Conference on International Language Resources and Evaluation (LREC), 2010, pp. 1320-1326.
  44. Pang, B., L. Lee, and S. Vaithyanathan, "Thumbs up? Sentiment classification using machine learning techniques", Proceedings of Annual Meeting of the Association for Computational Linguistics (ACL-2002), 2002.
  45. Paradarami, T. K., N. D. Bastian, and J. L. Wightman, "A hybrid recommender system using artificial neural networks", Expert Systems with Applications, Vol.83, 2017, pp. 300-313. https://doi.org/10.1016/j.eswa.2017.04.046
  46. Park, D. H., H. K. Kim, I. Y. Choi, and J. K. Kim, "A literature review and classification of recommender systems research", Expert Systems with Applications, Vol.39, No.11, 2012, pp. 10059-10072. https://doi.org/10.1016/j.eswa.2012.02.038
  47. Park, K., J. Lee, and J. Choi, "Deep neural networks for news recommendations", Proceedings of the 2017 ACM on Conference on Information and Knowledge Management, 2017, pp. 2255-2258
  48. Ricci, F., L. Rokach and B. Shapira, Introduction to Recommender Systems Handbook, Springer, Boston, MA, 2011.
  49. Sarwar, B., G. Karypis, J. Konstan, and J. Riedl, "Application of dimensionality reduction in recommender system-a case study", ACM WebKDD 2000 Workshop, 2000.
  50. Seo, S., J. Huang, H. Yang, and Y. Liu, "Interpretable Convolutional Neural Networks with Dual Local and Global Attention for Review Rating Prediction", Proceedings of the Eleventh ACM Conference on Recommender Systems, 2017, pp. 297-305
  51. Severyn, A. and A. Moschitti, "Twitter sentiment analysis with deep convolutional neural networks", Proceedings of the 38th international ACM SIGIR Conference on Research and Development in Information Retrieval, 2015, pp. 959-962.
  52. Shen, T., T. Zhou, G. Long, J. Jiang, S. Pan, and C. Zhang, "Disan: Directional self-attention network for rnn/cnn-free language understanding", Proceedings of the AAAI Conference on Artificial Intelligence, 2018.
  53. Su, X. and T. M. Khoshgoftaar, "A survey of collaborative filtering techniques", Advances in Artificial Intelligence, Vol.2009, 2009, pp. 1-19 https://doi.org/10.1155/2009/421425
  54. Tang, J., H. Gao, X. Hu, and H. Liu, "Context-aware review helpfulness rating prediction", Proceedings of the 7th ACM Conference on Recommender Systems, 2013, pp. 1-8.
  55. Tai, K. S., R. Socher, and C. D. Manning, "Improved semantic representations from tree-structured long short-term memory networks", Proceeding of IJCNLP, 2015, pp. 1556-1566.
  56. Wen, Y., W. Zhang, R. Luo, and J. Wang, "Learning text representation using recurrent convolutional neural network with highway layers", Proceedings of the 39th ACM SIGIR Workshop on Neural Information Retrieval, 2016.
  57. Wielgosz, M., A. Skoczen, and M. Mertik, "Using LSTM recurrent neural networks for monitoring the LHC superconducting magnets", Nucl Instrum Methods A, Vol.867, 2017, pp. 40-50. https://doi.org/10.1016/j.nima.2017.06.020
  58. Yenter, A. and A. Verma, "Deep CNN-LSTM with combined kernels from multiple branches for IMDb review sentiment analysis", 2017 IEEE 8th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference (UEMCON), 2017, pp. 540-546.
  59. Zhang, L., S. Wang, and B. Liu, "Deep learning for sentiment analysis: A survey", Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, Vol.8, No.4, 2018, pp. e1253.
  60. Zhang, S., L. Yao, A. Sun, and Y. Tay, "Deep learning based recommender system: A survey and new perspectives", ACM Computing Surveys (CSUR), Vol.52, No.1, 2019, pp. 1-38. https://doi.org/10.1145/3285029
  61. Zhang, Y. and B. Wallace, "A sensitivity analysis of (and practitioners' guide to) convolutional neural networks for sentence classification", arXiv preprint arXiv:1510.03820, 2015.
  62. Zhang, Z., D. Zhang, and J. Lai, "urCF: User review enhanced collaborative filtering", 20th Americas Conference on Information Systems, 2014.
  63. Zheng, L., V. Noroozi, and S. Yu, "Joint deep modeling of users and items using reviews for recommendation", Proceedings of the Tenth ACM International Conference on Web Search and Data Mining, 2017, pp. 425-434.