Journal of Intelligence and Information Systems (지능정보연구)

Korea Intelligent Information System Society (한국지능정보시스템학회)
권호 표지
DOI QR코드

DOI QR Code

A Study on the Enhancing Recommendation Performance Using the Linguistic Factor of Online Review based on Deep Learning Technique

딥러닝 기반 온라인 리뷰의 언어학적 특성을 활용한 추천 시스템 성능 향상에 관한 연구

  • Dongsoo Jang (Department of Big Data Analytics, Kyung Hee University) ;
  • Qinglong Li (Department of Big Data Analytics, Kyung Hee University) ;
  • Jaekyeong Kim (School of Management & Department of Big Data Analytics, Kyung Hee University)
  • 장동수 (경희대학교 대학원 빅데이터응용학과) ;
  • 이청용 (경희대학교 대학원 빅데이터응용학과) ;
  • 김재경 (경희대학교 경영대학&빅데이터응용학과)
  • Received : 2022.10.18
  • Accepted : 2022.12.07
  • Published : 2023.03.31
Download PDF
⟨ Previous Next ⟩

Abstract

As the online e-commerce market growing, the need for a recommender system that can provide suitable products or services to customer is emerging. Recently, many studies using the sentiment score of online review have been proposed to improve the limitations of study on recommender systems that utilize only quantitative information. However, this methodology has limitation in extracting specific preference information related to customer within online reviews, making it difficult to improve recommendation performance. To address the limitation of previous studies, this study proposes a novel recommendation methodology that applies deep learning technique and uses various linguistic factors within online reviews to elaborately learn customer preferences. First, the interaction was learned nonlinearly using deep learning technique for the purpose to extract complex interactions between customer and product. And to effectively utilize online review, cognitive contents, affective contents, and linguistic style matching that have an important influence on customer's purchasing decisions among linguistic factors were used. To verify the proposed methodology, an experiment was conducted using online review data in Amazon.com, and the experimental results confirmed the superiority of the proposed model. This study contributed to the theoretical and methodological aspects of recommender system study by proposing a methodology that effectively utilizes characteristics of customer's preferences in online reviews.

전자상거래 시장의 꾸준한 성장으로 인해 추천 시스템의 필요성은 점차 강조되고 있으며, 최근에는 추천 성능의 향상을 목적으로 리뷰 텍스트를 사용하는 연구가 활발히 진행되고 있다. 특히 많은 연구들은 리뷰 텍스트의 감성 점수를 활용하여 제안되고 있는데, 감성 점수만을 사용하는 방법론은 리뷰 텍스트에 존재하는 구체적인 선호도 정보의 활용 측면에 한계를 가지며 이는 결과적으로 성능 향상에 제약으로 작용하게 된다. 이를 개선하기 위해 본 연구는 딥러닝 기반 추천 모델에 온라인 리뷰 내 다양한 언어학적 요소들을 활용하여 고객의 선호도를 정교하게 학습할 수 있는 새로운 추천 방법론을 제안하였다. 이를 위해 먼저 고객과 상품 간 복잡한 상호작용을 고려할 수 있도록 딥러닝 모델을 통해 상호작용 관계를 비선형으로 학습하였다. 그리고 리뷰 텍스트를 효과적으로 활용할 수 있도록 언어학적 요소 중 고객의 구매 의사결정에 중요한 영향을 미치는 인지적 요인, 정서적 요인 그리고 언어 스타일 매칭을 사용하였다. 실험은 Amazon.com에서 수집한 온라인 리뷰 데이터를 사용하여 진행하였고, 실험 결과 제안 모델의 우수함을 검증할 수 있었다. 본 연구는 추천 시스템에서 리뷰 텍스트 내 고객 선호도에 대한 정보를 효과적으로 활용하는 방법론을 제안하여 연구의 이론적 및 방법론 측면에 기여하였다.

Keywords

Acknowledgement

본 논문은 교육부 및 한국연구재단 4단계 두뇌한국21 사업(4단계 BK21 사업)으로부터 지원받은 연구임.

References

  1. 김동언, 김민지, 김재경. (2022). 소규모 전자상거래를 위한 추천 시스템의 시간 차이에 따른 추천 효과 측정에 관한 연구. 인터넷전자상거래연구, 22(6), 185-202.
  2. 이승우, 강경모, 이병현, 이청용, 김재경. (2022). 사용자의 정성적 선호도와 정량적 선호도를 고려하는 추천 시스템 성능 향상에 관한 연구. 경영과학, 39(1), 15-27.
  3. 이청용, 이병현, 이흠철, 김재경. (2021). CNN 기반 리뷰 유용성 점수 예측을 통한 개인화 추천 서비스 성능 향상에 관한 연구. 지능정보연구, 27(3), 29-56. https://doi.org/10.13088/JIIS.2021.27.3.029
  4. 이흠철, 윤효림, 이청용, 김재경. (2022). Multi-channel CNN 기반 온라인 리뷰 유용성 예측 모델 개발에 관한 연구. 지능정보연구, 28(2), 171-189. https://doi.org/10.13088/JIIS.2022.28.2.171
  5. 전병국, 안현철. (2015). 사용자 리뷰 마이닝을 결합한 협업 필터링 시스템: 스마트폰 앱 추천에의 응용. 지능정보연구, 21(2), 1-18. https://doi.org/10.13088/JIIS.2015.21.2.01
  6. 현지연, 유상이, 이상용. (2019). 평점과 리뷰 텍스트 감성분석을 결합한 추천시스템 향상 방안 연구. 지능정보연구, 25(1), 219-239. https://doi.org/10.13088/JIIS.2019.25.1.219
  7. Acilar, A. M., & Arslan, A. (2009). A collaborative filtering method based on artificial immune network. Expert Systems with Applications, 36(4), 8324-8332. https://doi.org/10.1016/j.eswa.2008.10.029
  8. Alashkar, T., Jiang, S., Wang, S., & Fu, Y. (2017). Examples-rules guided deep neural network for makeup recommendation. Proceedings of the AAAI Conference on Artificial Intelligence, 31(1).
  9. Asani, E., Vahdat-Nejad, H., & Sadri, J. (2021). Restaurant recommender system based on sentiment analysis. Machine Learning with Applications, 6, 100114.
  10. Bennett, J., & Lanning, S. (2007). The netflix prize. Proceedings Of KDD Cup and Workshop, 35.
  11. Bhojne, N., Deore, S., Jagtap, R., Jain, G., & Kalal, C. (2017). Collaborative approach based restaurant recommender system using naive bayes. International Journal of Advanced Research in Computer and Communication Engineering, 6(4), 6-13.
  12. Billsus, D., & Pazzani, M. J. (1998). Learning collaborative information filters. Proceedings of the Fifteenth International Conference on Machine Learning, 46-54.
  13. Cao, R., Zhang, X., & Wang, H. (2019). A review semantics based model for rating prediction. IEEE Access, 8, 4714-4723. https://doi.org/10.1109/access.2019.2962075
  14. Chen, C., Zhao, P., Li, L., Zhou, J., Li, X., & Qiu, M. (2017). Locally connected deep learning framework for industrial-scale recommender systems. Proceedings of the 26th International Conference on World Wide Web Companion, 769-770.
  15. Cheng, Z., Ding, Y., Zhu, L., & Kankanhalli, M. (2018). Aspect-aware latent factor model: rating prediction with ratings and reviews. Proceedings of the 2018 World Wide Web Conference, 23-27.
  16. Chung, C., & Pennebaker, J. W. (2007). The psychological functions of function words. Social Communication, 1, 343-359.
  17. Das, A. S., Datar, M., Garg, A., & Rajaram, S. (2007). Google news personalization: scalable online collaborative filtering. Proceedings of the 16th International Conference on World Wide Web, 271-280.
  18. Garcia-Cumbreras, M. A., Montejo-Raez, A., & Diaz-Galiano, M. C. (2013). Pessimists and optimists: Improving collaborative filtering through sentiment analysis. Expert Systems with Applications, 40(17), 6758-6765. https://doi.org/10.1016/j.eswa.2013.06.049
  19. Goldberg, D., Nichols, D., Oki, B. M., & Terry, D. (1992). Using collaborative filtering to weave an information tapestry. Communications of the ACM, 35(12), 61-70. https://doi.org/10.1145/138859.138867
  20. Gonzales, A. L., Hancock, J. T., & Pennebaker, J. W. (2010). Language style matching as a predictor of social dynamics in small groups. Communication Research, 37(1), 3-19. https://doi.org/10.1177/0093650209351468
  21. Gottschalk, L. A. (2020). Text Analysis for The Social Sciences: The Unobtrusive Measurement of Psychological States and Traits. New York, United States: Routledge.
  22. Gundecha, P., & Liu, H. (2012). Mining social media: a brief introduction. New Directions in Informatics, Optimization, Logistics, and Production, 1-17.
  23. Haucap, J., & Heimeshoff, U. (2014). Google, Facebook, Amazon, eBay: Is the Internet driving competition or market monopolization? International Economics and Economic Policy, 11(1), 49-61. https://doi.org/10.1007/s10368-013-0247-6
  24. Hayati, H., Chanaa, A., Idrissi, M. K., & Bennani, S. (2019). Doc2Vec & Naive bayes: learners' cognitive presence assessment through asynchronous online discussion TQ transcripts. International Journal of Emerging Technologies in Learning, 14(8), 70-81.
  25. He, X., Liao, L., Zhang, H., Nie, L., Hu, X., & Chua, T.-S. (2017). Neural collaborative filtering. Proceedings of the 26th International Conference on World Wide Web, 173-182.
  26. Hickman, L., Thapa, S., Tay, L., Cao, M., & Srinivasan, P. (2022). Text preprocessing for text mining in organizational research: Review and recommendations. Organizational Research Methods, 25(1), 114-146.
  27. Huang, Y.-F., & Chen, P.-H. (2020). Fake news detection using an ensemble learning model based on self-adaptive harmony search algorithms. Expert Systems with Applications, 159, 113584.
  28. Huang, Y., Liu, H., Li, W., Wang, Z., Hu, X., & Wang, W. (2020). Lifestyles in Amazon: Evidence from online reviews enhanced recommender system. International Journal of Market Research, 62(6), 689-706. https://doi.org/10.1177/1470785319844146
  29. Huang, Z., Chung, W., & Chen, H. (2004). A graph model for E-commerce recommender systems. Journal of the American Society for Information Science and Technology, 55(3), 259-274. https://doi.org/10.1002/asi.10372
  30. Ireland, M. E., & Pennebaker, J. W. (2010). Language style matching in writing: synchrony in essays, correspondence, and poetry. Journal of Personality and Social Psychology, 99(3), 549.
  31. Janke, J., Castelli, M., & Popovic, A. (2019). Analysis of the proficiency of fully connected neural networks in the process of classifying digital images. Benchmark of different classification algorithms on high-level image features from convolutional layers. Expert Systems with Applications, 135, 12-38. https://doi.org/10.1016/j.eswa.2019.05.058
  32. Jeong, S.-Y., & Kim, Y.-K. (2021). Deep learning-based context-aware recommender system considering contextual features. Applied Sciences, 12(1), 45.
  33. Kang, S., Hwang, J., Kweon, W., & Yu, H. (2020). DE-RRD: A knowledge distillation framework for recommender system. Proceedings of the 29th ACM International Conference on Information & Knowledge Management, 605-614.
  34. Kim, H.-N., Ji, A.-T., Ha, I., & Jo, G.-S. (2010). Collaborative filtering based on collaborative tagging for enhancing the quality of recommendation. Electronic Commerce Research and Applications, 9(1), 73-83. https://doi.org/10.1016/j.elerap.2009.08.004
  35. Kim, K.-j., & Ahn, H. (2008). A recommender system using GA K-means clustering in an online shopping market. Expert Systems with Applications, 34(2), 1200-1209. https://doi.org/10.1016/j.eswa.2006.12.025
  36. Kim, K.-W., & Park, D.-H. (2018). Individual thinking style leads its emotional perception: Development of web-style design evaluation model and recommendation algorithm depending on consumer regulatory focus. Journal of Intelligence and Information Systems, 24(4), 171-196.
  37. Kiran, R., Kumar, P., & Bhasker, B. (2020). DNNRec: A novel deep learning based hybrid recommender system. Expert Systems with Applications, 144, 113054.
  38. Lee, R.-K., Chung, N., & Hong, T. (2019). Developing the online reviews based recommender models for multi-attributes using deep learning. The Journal of Information Systems, 28(1), 97-114.
  39. Leung, C. W., Chan, S. C., & Chung, F.-l. (2006). Integrating collaborative filtering and sentiment analysis: A rating inference approach. Proceedings of the ECAI 2006 Workshop on Recommender Systems, 62-66.
  40. Li, Q., & Kim, J. (2021). A deep learning-based course recommender system for sustainable development in education. Applied Sciences, 11(19), 8993.
  41. Li, X., Wang, M., & Liang, T.-P. (2014). A multi-theoretical kernel-based approach to social network-based recommendation. Decision Support Systems, 65, 95-104.
  42. Li, Y., Lu, L., & Xuefeng, L. (2005). A hybrid collaborative filtering method for multiple-interests and multiple-content recommendation in E-Commerce. Expert Systems with Applications, 28(1), 67-77. https://doi.org/10.1016/j.eswa.2004.08.013
  43. Liu, S., Gao, C., Chen, Y., Jin, D., & Li, Y. (2021). Learnable embedding sizes for recommender systems. Proceedings of the 9th International Conference on Learning Representations (ICLR 2021). https://openreview.net/group?id=ICLR.cc/2021/Conference
  44. Ludwig, S., De Ruyter, K., Friedman, M., Bruggen, E. C., Wetzels, M., & Pfann, G. (2013). More than words: The influence of affective content and linguistic style matches in online reviews on conversion rates. Journal of Marketing, 77(1), 87-103. https://doi.org/10.1509/jm.11.0560
  45. Ma, Y., Chen, G., & Wei, Q. (2017). Finding users preferences from large-scale online reviews for personalized recommendation. Electronic Commerce Research, 17(1), 3-29.
  46. Nguyen, T. T., Hui, P.-M., Harper, F. M., Terveen, L., & Konstan, J. A. (2014). Exploring the filter bubble: the effect of using recommender systems on content diversity. Proceedings of the 23rd International Conference on World Wide Web, 677-686.
  47. Ni, J., Li, J., & McAuley, J. (2019). Justifying recommendations using distantly-labeled reviews and fine-grained aspects. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, 188-197.
  48. Ning, X., & Karypis, G. (2011). Slim: Sparse linear methods for top-n recommender systems. Proceedings of the 2011 IEEE 11th International Conference on Data Mining, 497-506.
  49. Park, Y., Park, S., Jung, W., & Lee, S.-g. (2015). Reversed CF: A fast collaborative filtering algorithm using a k-nearest neighbor graph. Expert Systems with Applications, 42(8), 4022-4028. https://doi.org/10.1016/j.eswa.2015.01.001
  50. Preethi, G., Krishna, P. V., Obaidat, M. S., Saritha, V., & Yenduri, S. (2017). Application of deep learning to sentiment analysis for recommender system on cloud. Proceedings of the 2017 International Conference on Computer, 93-97.
  51. Rama, K., Kumar, P., & Bhasker, B. (2021). Deep autoencoders for feature learning with embeddings for recommendations: a novel recommender system solution. Neural Computing and Applications, 33(21), 14167-14177. https://doi.org/10.1007/s00521-021-06065-9
  52. Rendle, S. (2012a). Factorization machines with libfm. ACM Transactions on Intelligent Systems and Technology (TIST), 3(3), 1-22. https://doi.org/10.1145/2168752.2168771
  53. Rendle, S. (2012b). Learning recommender systems with adaptive regularization. Proceedings of the Fifth ACM International Conference on Web Search and Data Mining, 133-142.
  54. Rendle, S., & Schmidt-Thieme, L. (2008). Online-updating regularized kernel matrix factorization models for large-scale recommender systems. Proceedings of the 2008 ACM Conference on Recommender Systems, 251-258.
  55. Roberts, C. W. (2020). Text Analysis for the Social Sciences: Methods for Drawing Statistical Inferences from Texts and Transcripts. New York, United States: Routledge.
  56. Salsabila, G. D., & Setiawan, E. B. (2021). Semantic Approach for Big Five Personality Prediction on Twitter. Jurnal RESTI, 5(4), 680-687.
  57. Sarwar, B., Karypis, G., Konstan, J., & Riedl, J. (2001). Item-based collaborative filtering recommendation algorithms. Proceedings of the 10th international conference on World Wide Web, 285-295.
  58. Schafer, J. B., Konstan, J. A., & Riedl, J. (2001). E-commerce recommendation applications. Data Mining and Knowledge Discovery, 5(1), 115-153. https://doi.org/10.1023/A:1009804230409
  59. Silveira, T., Zhang, M., Lin, X., Liu, Y., & Ma, S. (2019). How good your recommender system is? A survey on evaluations in recommendation. International Journal of Machine Learning and Cybernetics, 10(5), 813-831. https://doi.org/10.1007/s13042-017-0762-9
  60. Srivastava, N. (2013). Improving neural networks with dropout. University of Toronto.
  61. Su, X., & Khoshgoftaar, T. M. (2009). A survey of collaborative filtering techniques. Advances in Artificial Intelligence, 4, 2.
  62. Tausczik, Y. R., & Pennebaker, J. W. (2010). The psychological meaning of words: LIWC and computerized text analysis methods. Journal of Language and Social Psychology, 29(1), 24-54. https://doi.org/10.1177/0261927X09351676
  63. Topaloglu, O., & Dass, M. (2021). The impact of online review content and linguistic style matching on new product sales: The moderating role of review helpfulness. Decision Sciences, 52(3), 749-775. https://doi.org/10.1111/deci.12378
  64. Unger, M., Tuzhilin, A., & Livne, A. (2020). Context-aware recommendations based on deep learning frameworks. ACM Transactions on Management Information Systems (TMIS), 11(2), 1-15. https://doi.org/10.1145/3386243
  65. Wang, H., Amagata, D., Maekawa, T., Hara, T., Niu, H., Yonekawa, K., & Kurokawa, M. (2019). Preliminary investigation of alleviating user cold-start problem in e-commerce with deep cross-domain recommender system. Proceedings of the 2019 World Wide Web Conference, 398-403
  66. Wang, X., Lin, X., & Spencer, M. K. (2019). Exploring the effects of extrinsic motivation on consumer behaviors in social commerce: Revealing consumers' perceptions of social commerce benefits. International Journal of Information Management, 45, 163-175. https://doi.org/10.1016/j.ijinfomgt.2018.11.010
  67. Wang, X., Tang, L. R., & Kim, E. (2019). More than words: Do emotional content and linguistic style matching matter on restaurant review helpfulness? International Journal of Hospitality Management, 77, 438-447.
  68. Wang, Y., Wang, M., & Xu, W. (2018). A sentiment-enhanced hybrid recommender system for movie recommendation: a big data analytics framework. Wireless Communications and Mobile Computing. https://doi.org/10.1155/2018/8263704
  69. Yang, J., Yi, X., Zhiyuan Cheng, D., Hong, L., Li, Y., Xiaoming Wang, S., Xu, T., & Chi, E. H. (2020). Mixed negative sampling for learning two-tower neural networks in recommendations. Proceedings of the Web Conference 2020, 441-447.
  70. Yang, S., Zhou, C., & Chen, Y. (2021). Do topic consistency and linguistic style similarity affect online review helpfulness? An elaboration likelihood model perspective. Information Processing & Management, 58(3), 102521.
  71. Zeiler, M. D. (2012). Adadelta: an adaptive learning rate method. arXiv preprint arXiv:1212.5701. https://arxiv.org/abs/1212.5701
  72. Zhang, Z., Zhang, D., & Lai, J. (2014). urCF: user review enhanced collaborative filtering, University of Maryland.
  73. Zheng, L., Noroozi, V., & Yu, P. S. (2017). Joint deep modeling of users and items using reviews for recommendation. Proceedings of the Tenth ACM International Conference on Web Search and Data Mining, 425-434.
  74. Zhou, X., He, J., Huang, G., & Zhang, Y. (2015). SVD-based incremental approaches for recommender systems. Journal of Computer and System Sciences, 81(4), 717-733. https://doi.org/10.1016/j.jcss.2014.11.016