|
|
|
| Research on the Post-hoc Explanation Recommendation Model Based on Generative AI |
| Li Weiqing1,4, Wang Weijun2, Huang Yinghui3, Huang Wei1, Zhang Rui1 |
1.School of Economics and Management, Hubei University of Technology, Wuhan 430068
2.Key Laboratory of Adolescent Cyberpsychology and Behavior (Central China Normal University), Ministry of Education, Wuhan 430079
3.School of Management, Wuhan University of Technology, Wuhan 430070
4.Hubei Research Center for Digital Industrial Economy Development, Wuhan 430068 |
|
|
|
|
Abstract We propose a post-hoc explanation recommendation model based on generative artificial intelligence (GenAI), integrating the theory of consumption values into the recommendation process. This approach enhances the recommendation effectiveness while generating personalized textual explanations for users. Using generative AI prompt engineering, we assess fine-grained consumption value tendencies—functional, symbolic, economic, and emotional—extracted from product reviews. Based on these assessments, a user-item (value) preference interaction matrix is constructed, enabling the implementation of a value-driven recommendation model. Finally, the recommendation results and corresponding consumption value tendencies are fed into a generative AI-based explanation bot to produce tailored and contextually relevant explanations. The results demonstrate that the generative AI-based scoring bot achieves high accuracy, consistency, and diversity in evaluating consumption value tendencies, providing critical support for both the recommendation model and explanation system. By incorporating consumption values, the proposed model significantly improves recommendation accuracy and diversity while excelling in cold start and data sparsity scenarios. This approach offers new insights into addressing issues such as the “filter bubble” and over-specialization in recommendation systems. Additionally, the explanations generated by the generative AI-based bot are fluent, logically coherent, and can effectively communicate the underlying recommendation mechanisms and value tendencies. Compared with traditional methods, this approach is more flexible, efficient, and personalized, offering a novel solution to enhance the recommendation system transparency and foster user trust.
|
|
Received: 14 January 2025
|
|
|
|
1 Zhou X F, Zhou Y S, Gong Y F, et al. “Bee and I need diversity!” Break filter bubbles in recommendation systems through embodied AI learning[C]// Proceedings of the 23rd Annual ACM Interaction Design and Children Conference. New York: ACM Press, 2024: 44-61.
2 姜婷婷, 许艳闰. 国外过滤气泡研究: 基础、脉络与展望[J]. 情报学报, 2021, 40(10): 1108-1117.
3 Zhang Y F, Chen X. Explainable recommendation: a survey and new perspectives[J]. Foundations and Trends? in Information Retrieval, 2020, 14(1): 1-101.
4 高广尚. 可解释推荐模型中的可解释性方法研究综述[J]. 数据分析与知识发现, 2024, 8(8/9): 6-19.
5 Tintarev N. Explanations of recommendations[C]// Proceedings of the 2007 ACM Conference on Recommender Systems. New York: ACM Press, 2007: 203-206.
6 Turbé H, Bjelogrlic M, Lovis C, et al. Evaluation of post-hoc interpretability methods in time-series classification[J]. Nature Machine Intelligence, 2023, 5(3): 250-260.
7 Chakraborty D, Paul J. Healthcare apps’ purchase intention: a consumption values perspective[J]. Technovation, 2023, 120: 102481.
8 杨珂欣, 张奇, 余乐安, 等. 基于消费者价值观和有限理性的新能源汽车购买意愿与助推政策研究[J]. 管理评论, 2023, 35(1): 146-158.
9 Chang Y P, Wang X, Wang J D, et al. A survey on evaluation of large language models[J]. ACM Transactions on Intelligent Systems and Technology, 2024, 15(3): Article No.39.
10 罗震, 张涛. 融合用户特征与知识图谱的可解释电影推荐方法[J]. 管理工程学报, 2024, 38(6): 128-139.
11 Chatti M A, Guesmi M, Muslim A. Visualization for recommendation explainability: a survey and new perspectives[J]. ACM Transactions on Interactive Intelligent Systems, 2024, 14(3): Article No.19.
12 李伟卿, 王伟军, 黄炜, 等. 可解释信息推荐研究综述[J]. 情报学报, 2023, 42(7): 870-882.
13 Retzlaff C O, Angerschmid A, Saranti A, et al. Post-hoc vs ante-hoc explanations: xAI design guidelines for data scientists[J]. Cognitive Systems Research, 2024, 86: 101243.
14 Guidotti R, Monreale A, Ruggieri S, et al. A survey of methods for explaining black box models[J]. ACM Computing Surveys, 2019, 51(5): Article No.93.
15 Rehman Zafar M, Khan N M. DLIME: a deterministic local interpretable model-agnostic explanations approach for computer-aided diagnosis systems[OL]. (2019-06-24). https://arxiv.org/pdf/1906.10263.
16 Brunot L, Canovas N, Chanson A, et al. Preference-based and local post-hoc explanations for recommender systems[J]. Information Systems, 2022, 108: 102021.
17 Zhong J F, Negre E. Shap-enhanced counterfactual explanations for recommendations[C]// Proceedings of the 37th ACM/SIGAPP Symposium on Applied Computing. New York: ACM Press, 2022: 1365-1372.
18 Le N L, Abel M H, Gouspillou P. Combining embedding-based and semantic-based models for post-hoc explanations in recommender systems[C]// Proceedings of the 2023 IEEE International Conference on Systems, Man, and Cybernetics. Piscataway: IEEE, 2023: 4619-4624.
19 Wang X T, Chen Y R, Yang J, et al. A reinforcement learning framework for explainable recommendation[C]// Proceedings of the 2018 IEEE International Conference on Data Mining. Piscataway: IEEE, 2018: 587-596.
20 Wang X M, Li Q, Yu D E, et al. Reinforced path reasoning for counterfactual explainable recommendation[J]. IEEE Transactions on Knowledge and Data Engineering, 2024, 36(7): 3443-3459.
21 Amin S, Tarun M T. Effect of consumption values on customers’ green purchase intention: a mediating role of green trust[J]. Social Responsibility Journal, 2021, 17(8): 1320-1336.
22 Gutman J. A means-end chain model based on consumer categorization processes[J]. Journal of Marketing, 1982, 46(2): 60-72.
23 董雅丽, 杜振涛, 李治国. 消费价值观对消费行为影响的实证研究[J]. 调研世界, 2011(10): 47-50.
24 Adeola O, Moradeyo A A, Muogboh O, et al. Consumer values, online purchase behaviour and the fashion industry: an emerging market context[J]. PSU Research Review, 2024, 8(1): 68-83.
25 Alavi M, Leidner D E, Mousavi R. Knowledge management perspective of generative artificial intelligence[J]. Journal of the Association for Information Systems, 2024, 25(1): 1-12.
26 Zhao W X, Zhou K, Li J Y, et al. A survey of large language models[OL]. (2025-03-11). https://arxiv.org/pdf/2303.18223.
27 Demszky D, Yang D Y, Yeager D S, et al. Using large language models in psychology[J]. Nature Reviews Psychology, 2023, 2(11): 688-701.
28 Rathje S, Mirea D M, Sucholutsky I, et al. GPT is an effective tool for multilingual psychological text analysis[J]. Proceedings of the National Academy of Sciences of the United States of America, 2024, 121(34): e2308950121.
29 Biró A, Cuesta-Vargas A I, Szilágyi L. Precognition of mental health and neurogenerative disorders using AI-parsed text and sentiment analysis[J]. Acta Universitatis Sapientiae, Informatica, 2023, 15(2): 359-403.
30 Chen B Y, Wu Z X, Zhao R R. From fiction to fact: the growing role of generative AI in business and finance[J]. Journal of Chinese Economic and Business Studies, 2023, 21(4): 471-496.
31 Zhang Q, Wang Y L, Ariffin S K.Consumers purchase intention in live-streaming e-commerce:a consumption value perspective and the role of streamer popularity[J]. PLoS One, 2024, 19(2):e0296339.
32 Joshi Y, Uniyal D P, Sangroya D. Investigating consumers’ green purchase intention: examining the role of economic value, emotional value and perceived marketplace influence[J]. Journal of Cleaner Production, 2021, 328: 129638.
33 Tandon A, Kaur P, Bhatt Y, et al. Why do people purchase from food delivery apps? A consumer value perspective[J]. Journal of Retailing and Consumer Services, 2021, 63: 102667.
34 Zhang N, Liu R, Zhang X Y, et al. The impact of consumer perceived value on repeat purchase intention based on online reviews: by the method of text mining[J]. Data Science and Management, 2021, 3: 22-32.
35 Stray J, Halevy A, Assar P, et al. Building human values into recommender systems: an interdisciplinary synthesis[J]. ACM Transactions on Recommender Systems, 2024, 2(3): Article No.20.
36 Zhao Y Y, Wang Y, Liu Y C, et al. Fairness and diversity in recommender systems: a survey[J]. ACM Transactions on Intelligent Systems and Technology, 2025, 16(1): Article No.2.
37 李伟卿, 池毛毛, 王伟军. 面向用户长短期偏好调节的可解释个性化推荐方法研究[J]. 图书情报工作, 2021, 65(12): 101-111.
38 Jin B B, Lian D F, Liu Z, et al. Sampling-decomposable generative adversarial recommender[C]// Proceedings of the 34th Conference on Neural Information Processing Systems. Cham: Springer, 2020: 22629-22639.
39 Li J H, Li J H, Wang C J, et al. Wide & deep generative adversarial networks for recommendation system[J]. Intelligent Data Analysis, 2023, 27(1): 121-136.
40 Wu L K, Zheng Z, Qiu Z P, et al. A survey on large language models for recommendation[J]. World Wide Web, 2024, 27(5): Article No.60.
41 Deldjoo Y, He Z K, McAuley J, et al. Recommendation with generative models[OL]. (2024-09-18). https://arxiv.org/pdf/2409.15173.
42 Zhong W L, Jin R, Yang C, et al. Stock constrained recommendation in tmall[C]// Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM Press, 2015: 2287-2296. |
|
|
|
