Description-based Post-hoc Explanation for Twitter List Recommendations

Computer and Knowledge Engineering

مقاله 4، دوره 7، شماره 2 - شماره پیاپی 14، اسفند 2024، صفحه 43-50   

نوع مقاله: Original Article

شناسه دیجیتال (DOI): 10.22067/cke.2024.85185.1107

نویسندگان

Havva Alizadeh Noughabi؛ Behshid Behkamal^* ؛ Mohsen Kahani

Department of Computer Engineering, Ferdowsi University of Mashhad, Mashhad, Iran.

چکیده

Twitter List recommender systems can generate highly accurate recommendations, but since they employ heterogeneous information of users and Lists and apply complex prediction models, they cannot provide easy understandable intrinsic explanations. To address this limitation, Twitter List descriptions can play a critical role in providing post-hoc explanations that help users make informed decisions. In this paper, we propose an explanation model to provide relevant and informative explanations for recommended Lists by automatically generating descriptions for Twitter Lists. The model selects the most informative tweets from a List as its description to inform users more with the recommended List that positively contributes to the user experience. More specifically, the explanation model incorporates three categories of features: content relevance features, tweet-specific features, and publisher’s authority features that are used in a learning to rank model to rank the List’s tweets in terms of their informativeness. By conducting experiments on a Twitter dataset, we have shown that the proposed model provides useful explanations for the Lists that are recommended to users, while upholding parity in recommendation performance.

کلیدواژه‌ها

Explainable recommender systems؛ Post-hoc explanation؛ Description generation؛ Twitter Lists

مراجع

[1]  S. de la Rouviere and K. Ehlers, “Lists as coping strategy for information overload on Twitter,” Proceedings of the 22nd International Conference on World Wide Web, 2013, pp. 199–200.

[2]  V. Rakesh, D. Singh, B. Vinzamuri, and C. K. Reddy, “Personalized recommendation of twitter lists using content and network information,” Proceedings of the International AAAI Conference on Web and Social Media, 2014, pp. 416–425.

[3]  L. Chen, Y. Zhao, S. Chen, H. Fang, C. Li, and M. Wang, “iplug: Personalized list recommendation in twitter,” Web Information Systems Engineering–WISE 2013: 14th International Conference, Nanjing, China, 2013, pp. 88–103.

[4]  C.-H. Tsai and P. Brusilovsky. (2020, Oct.). The effects of controllability and explainability in a social recommender system. User Modeling and User-Adapted Interaction. [Online]. 31, pp. 591–627. Available: 10.1007/s11257-020-09281-5

[5]  A. Papadimitriou, P. Symeonidis, and Y. Manolopoulos. (2012, May.). A generalized taxonomy of explanations styles for traditional and social recommender systems. Data Mining and Knowledge Discovery. [Online]. 24, pp. 555–583. Available: 10.1007/s10618-011-0215-0

[6]  Y. Zhang and X. Chen. (2020, Mar.). Explainable recommendation: A survey and new perspectives. Foundations and Trends® in Information Retrieval. [Online]. 14(1), pp. 1–101. Available: 10.1561/1500000066

[7]  M. T. Ribeiro, S. Singh, and C. Guestrin, “Model-agnostic interpretability of machine learning,” arXiv preprint arXiv:1606.05386, 2016.

[8]  D. Shmaryahu, G. Shani, and B. Shapira, “Post-hoc Explanations for Complex Model Recommendations using Simple Methods,” IntRS@ RecSys, 2020, pp. 26–36.

[9]  Q. Chen, J. Lin, Y. Zhang, H. Yang, J. Zhou, and J. Tang, “Towards knowledge-based personalized product description generation in e-commerce,” Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2019, pp. 3040–3050.

[10] A. Nadamoto, K. Fukumoto, R. Takeuchi, H. Terada, and M. Bato. (2023). Automatic generation of product description using deep learning methods. Journal of Data Intelligence. [Online]. 4(1 & 2), pp. 134–148. Available: rintonpress.com

[11]         Q. Zhang, B. Guo, S. Liu, J. Liu, and Z. Yu. (2022, Apr.). CrowdDesigner: information-rich and personalized product description generation. Frontiers of Computer Science. [Online]. 16(6), p. 166339. Available: 10.1007/s11704-022-1193-7

[12] J. Wang, Y. Hou, J. Liu, Y. Cao, and C.-Y. Lin, “A statistical framework for product description generation,” Proceedings of the Eighth International Joint Conference on Natural Language Processing, 2017, pp. 187–192.

[13] S. Gerani, Y. Mehdad, G. Carenini, R. Ng, and B. Nejat, “Abstractive summarization of product reviews using discourse structure,” Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP), 2014, pp. 1602–1613.

[14] K. Balog and F. Radlinski, “Measuring recommendation explanation quality: The conflicting goals of explanations,” Proceedings of the 43rd international ACM SIGIR conference on research and development in information retrieval, 2020, pp. 329–338.

[15]  I. Nunes and D. Jannach. (2017, Oct.). A systematic review and taxonomy of explanations in decision support and recommender systems. User Modeling and User-Adapted Interaction. [Online]. 27, pp. 393–444. Available: 10.1007/s11257-017-9195-0

[16] N. Tintarev and J. Masthoff, “A survey of explanations in recommender systems,” IEEE 23rd international conference on data engineering workshop, Istanbul, Turkey, 2007, pp. 801–810.

[17] M. Caro-Martínez, G. Jiménez-Díaz, and J. A. Recio-García. (2021, Jul.). Conceptual modeling of explainable recommender systems: an ontological formalization to guide their design and development. Journal of Artificial Intelligence Research. [Online]. 71, pp. 557–589. Available: 10.1613/jair.1.12789

[18] C. Nóbrega and L. Marinho, “Towards explaining recommendations through local surrogate models,” Proceedings of the 34th ACM/SIGAPP Symposium on Applied Computing, 2019, pp. 1671–1678.

[19] B. C. G. Lee, K. Lo, D. Downey, and D. S. Weld, “Explanation-based tuning of opaque machine learners with application to paper recommendation,” arXiv preprint arXiv:2003.04315, 2020.

[20]  G. Peake and J. Wang, “Explanation mining: Post hoc interpretability of latent factor models for recommendation systems,” Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2018, pp. 2060–2069.

[21]  C. Lonjarret, C. Robardet, M. Plantevit, R. Auburtin, and M. Atzmueller, “Why should i trust this item? explaining the recommendations of any model,” IEEE 7th International Conference on Data Science and Advanced Analytics (DSAA), Sydney, NSW, Australia, 2020, pp. 526–535.

[22]  M. Du, N. Liu, and X. Hu. (2019, Dec.). Techniques for interpretable machine learning. Communications of the ACM. [Online]. 63(1), pp. 68–77. Available: 10.1145/3359786

[23]  Y. Zhang, G. Lai, M. Zhang, Y. Zhang, Y. Liu, and S. Ma, “Explicit factor models for explainable recommendation based on phrase-level sentiment analysis,” 37th international ACM SIGIR conference on Research & development in information retrieval, 2014, pp. 83–92.

[24]  N. Wang, H. Wang, Y. Jia, and Y. Yin, “Explainable recommendation via multi-task learning in opinionated text data,” 41st International ACM SIGIR Conference on Research & Development in Information Retrieval, 2018, pp. 165–174.

[25]  Q. Ai, V. Azizi, X. Chen, and Y. Zhang. (2018, Sep.). Learning heterogeneous knowledge base embeddings for explainable recommendation. Algorithms. [Online]. 11(9), p. 137. Available: 10.3390/a11090137

[26]  H. Wang et al. and M. Guo, “Ripplenet: Propagating user preferences on the knowledge graph for recommender systems,” Proceedings of the 27th ACM international conference on information and knowledge management, 2018, pp. 417–426.

[27]  Y. Xian, Z. Fu, S. Muthukrishnan, G. De Melo, and Y. Zhang, “Reinforcement knowledge graph reasoning for explainable recommendation,” Proceedings of the 42nd international ACM SIGIR conference on research and development in information retrieval, 2019, pp. 285–294.

[28]  F. Fusco, M. Vlachos, V. Vasileiadis, K. Wardatzky, and J. Schneider, “RecoNet: An Interpretable Neural Architecture for Recommender Systems.,” International Joint Conference on Artificial Intelligence, 2019, pp. 2343–2349.

[29]  Y. Lu, R. Dong, and B. Smyth, “Why I like it: multi-task learning for recommendation and explanation,” Proceedings of the 12th ACM Conference on Recommender Systems, 2018, pp. 4–12.

[30]  Z. Chen et al. and E. Chen, “Co-attentive multi-task learning for explainable recommendation.,” International Joint Conference on Artificial Intelligence, 2019, pp. 2137–2143.

[31]  X. Wang, X. He, F. Feng, L. Nie, and T.-S. Chua, “Tem: Tree-enhanced embedding model for explainable recommendation,” Proceedings of the 2018 world wide web conference, 2018, pp. 1543–1552.

[32]  W. Ma et al. and X. Ren, “Jointly learning explainable rules for recommendation with knowledge graph,” The world wide web conference, 2019, pp. 1210–1221.

[33]  W. Sherchan, S. Nepal, and C. Paris. (2013, Aug.). A survey of trust in social networks. ACM Computing Surveys (CSUR). [Online]. 45(4), pp. 1–33. Available: 10.1145/2501654.250166

[34]  A. Sharma and D. Cosley, “Do social explanations work? Studying and modeling the effects of social explanations in recommender systems,” Proceedings of the 22nd international conference on World Wide Web, 2013, pp. 1133–1144.

[35]  B. Wang, M. Ester, J. Bu, and D. Cai, “Who also likes it? generating the most persuasive social explanations in recommender systems,” Proceedings of the AAAI Conference on Artificial Intelligence, 2014.

[36]  C. Shi, Z. Zhang, P. Luo, P. S. Yu, Y. Yue, and B. Wu, “Semantic path based personalized recommendation on weighted heterogeneous information networks,” Proceedings of the 24th ACM International on Conference on Information and Knowledge Management, 2015, pp. 453–462.

[37]  Y. Zhang, X. Xu, H. Zhou, and Y. Zhang, “Distilling structured knowledge into embeddings for explainable and accurate recommendation,” Proceedings of the 13th international conference on web search and data mining, 2020, pp. 735–743.

[38]  Z. Ren, S. Liang, P. Li, S. Wang, and M. de Rijke, “Social collaborative viewpoint regression with explainable recommendations,” Proceedings of the tenth ACM international conference on web search and data mining, 2017, pp. 485–494.

[39]   J. Zheng, Q. Li, J. Liao, and S. Wang. (2021, Oct.). Explainable link prediction based on multi-granularity relation-embedded representation. Knowledge-Based Systems. [Online]. 230, p. 107402. Available: 10.1016/j.knosys.2021.107402

[40]  J. Zheng, Z. Qin, S. Wang, and D. Li. (2022, Jun.). Attention-based explainable friend link prediction with heterogeneous context information. Information Sciences. [Online]. 597, pp. 211–229, 2022. Available: 10.1016/j.ins.2022.03.010

[41]  D. Rudrapal, A. Das, and B. Bhattacharya. (2019, Jun.). Ranking of event-focused english tweets based on relevance and informativeness. Computación y Sistemas. [Online]. 23(2), pp. 491–500. Available: 10.13053/cys-23-2-3011

[42]  T. Zhang, V. Kishore, F. Wu, K. Q. Weinberger, and Y. Artzi, “Bertscore: Evaluating text generation with bert,” arXiv preprint arXiv:1904.09675, 2019.

[43]  K. Chen, T. Chen, G. Zheng, O. Jin, E. Yao, and Y. Yu, “Collaborative personalized tweet recommendation,” Proceedings of the 35th international ACM SIGIR conference on Research and development in information retrieval, 2012, pp. 661–670.

[44]  C. De Maio, G. Fenza, M. Gallo, V. Loia, and M. Parente. (2019, Apr.). Time-aware adaptive tweets ranking through deep learning. Future Generation Computer Systems. [Online]. 93, pp. 924–932. Available: 10.1016/j.future.2017.07.039

[45]  Y. Duan, L. Jiang, T. Qin, M. Zhou, and H. Y. Shum, “An empirical study on learning to rank of tweets,” Proceedings of the 23rd international conference on computational linguistics (Coling 2010), 2010, pp. 295–303.

[46] K. Sailunaz, J. Kawash, and R. Alhajj. (2022, Apr.). Tweet and user validation with supervised feature ranking and rumor classification. Multimedia Tools and Applications. [Online]. 81(22), pp. 31907–31927. Available: 10.1007/s11042-022-12616-6

[47] D. Kim, Y. Jo, I.-C. Moon, and A. Oh, “Analysis of twitter lists as a potential source for discovering latent characteristics of users,” ACM CHI workshop on microblogging, Citeseer, 2010.

[48] D. Metzler and W. Bruce Croft. (2007, Jan.). Linear feature-based models for information retrieval. Information Retrieval. 10, pp. 257–274. Available: 10.1007/s10791-006-9019-z

[49] J. H. Friedman. (2001, Oct.). Greedy function approximation: a gradient boosting machine. Annals of statistics. [Online]. 29(5), pp. 1189–1232. Available: jstor.org/stable/2699986

[50] Q. Wu, C. J. Burges, K. M. Svore, and J. Gao. (2009, Sep.). Adapting boosting for information retrieval measures.  Information Retrieval. [Online]. 13, pp. 254–270. Available: 10.1007/s10791-009-9112-1

[51] C. Chen, M. Zhang, Y. Liu, and S. Ma, “Neural attentional rating regression with review-level explanations,” Proceedings of the 2018 world wide web conference, 2018, pp. 1583–1592.