Learning to rank aspects and opinions for comparative explanations
Abstract
Comparative recommendation explanations help to make sense of recommendations by comparing a recommended item along some aspects of interest with one or many items being considered. This work extends the notion of comparative explanations, by going beyond merely better/worse statements, to further incorporate aspect-level opinions for more informative comparisons. To enhance the quality of both the personalized recommendation and the explanation, we incorporate optimization objectives that preserve relative rankings of aspects and opinions, in addition to the classical rankings of overall preferences for items. We integrate the multiple ranking objectives and multi-tensor factorization together. Experiments on datasets of different domains validate the efficacy of our proposed framework in both recommendation and comparative explanation against comparable explainable recommendation baselines.
References
[1]
Amarouche K, Benbrahim H, and Kassou I Product opinion mining for competitive intelligence Procedia Computer Science 2015 73 358-365
[2]
Balloccu G, Boratto L, Fenu G, et al. Reinforcement recommendation reasoning through knowledge graphs for explanation path quality KBS 2023 260 110098
[3]
Bauman, K., Liu, B., & Tuzhilin, A. (2017). Aspect based recommendations: Recommending items with the most valuable aspects based on user reviews. In: KDD, KDD ’17.
[4]
Chen, C., Zhang, M., Liu, Y, et al. (2018). Neural attentional rating regression with review-level explanations. In: Proceedings of the 2018 World Wide Web Conference, WWW ’18, p 1583–1592.
[5]
Chen, H., Li, Y., Sun, X., et al. (2021). Temporal meta-path guided explainable recommendation. In: WSDM. ACM, WSDM ’21, p 1056–1064.
[6]
Chen, L. & Wang, F. (2017). Explaining recommendations based on feature sentiments in product reviews. In: IUI, IUI ’17, p 17–28.
[7]
Chen, X., Qin, Z., Zhang, Y., et al. (2016). Learning to rank features for recommendation over multiple categories. In: SIGIR. ACM, SIGIR ’16, p 305–314.
[8]
Chen, X., Chen, H., Xu, H., et al. (2019a). Personalized fashion recommendation with visual explanations based on multimodal attention network: Towards visually explainable recommendation. In: SIGIR, SIGIR’19.
[9]
Chen, X., Zhang, Y., & Qin, Z. (2019b). Dynamic explainable recommendation based on neural attentive models. In: AAAI, pp 53–60.
[10]
Chen, Z., Wang, X., Xie, X., et al. (2019c). Co-attentive multi-task learning for explainable recommendation. In: IJCAI, IJCAI ’19, pp 2137–2143.
[11]
Echterhoff, J., Yan, A., & McAuley, J. (2023). Comparing apples to apples: Generating aspect-aware comparative sentences from user review. arXiv preprint arXiv:2307.03691.
[12]
Feldman, R., Fresko, M., Goldenberg, J., et al. (2007). Extracting product comparisons from discussion boards. In: ICDM, IEEE, ICDM’07, pp 469–474.
[13]
Gao, J., Wang, X., Wang, Y., et al. (2019). Explainable recommendation through attentive multi-view learning. In: AAAI, pp 3622–3629.
[14]
Geng, S., Fu, Z., Tan, J., et al. (2022). Path language modeling over knowledge graphsfor explainable recommendation. In: WWW. ACM, New York, NY, USA, WWW ’22, p 946-955,.
[15]
Good IJ Maximum entropy for hypothesis formulation, especially for multidimensional contingency tables Annals of Mathematical Statistics 1963 34 3 911-934
[16]
Hada, DV., M. V, & Shevade, SK. (2021). Rexplug: Explainable recommendation using plug-and-play language model. In: SIGIR. ACM, New York, NY, USA, SIGIR ’21, p 81–91,.
[17]
He, R., & McAuley, J. (2016). Ups and downs: Modeling the visual evolution of fashion trends with one-class collaborative filtering. In: WWW, WWW ’16, p 507–517,.
[18]
He, X., Chen, T., Kan, MY., et al. (2015). Trirank: Review-aware explainable recommendation by modeling aspects. In: CIKM, CIKM ’15, p 1661–1670.
[19]
Herlocker JL, Konstan JA, Terveen LG, et al. Evaluating collaborative filtering recommender systems TOIS 2004 22 1 5-53
[20]
Kolda TG and Bader BW Tensor decompositions and applications SIAM Review 2009 51 3 455-500
[21]
Krippendorff, K. (2011). Computing krippendorff’s alpha-reliability.
[22]
Le, TH., & Lauw, HW. (2021). Explainable recommendation with comparative constraints on product aspects. In: WSDM, WSDM ’21, p 967–975.
[23]
Le, TH., & Lauw, HW. (2022). Question-attentive review-level recommendation explanation. In: BigData, IEEE, BigData’22, pp 756–761.
[24]
Li, L., Zhang, Y., & Chen, L. (2021). Personalized transformer for explainable recommendation. In: ACL, pp 4947–4957.
[25]
Li L, Zhang Y, and Chen L On the relationship between explanation and recommendation: Learning to rank explanations for improved performance ACM Transactions on Intelligent Systems and Technology 2023
[26]
Li, P., Wang, Z., Ren, Z., et al. (2017). Neural rating regression with abstractive tips generation for recommendation. In: SIGIR. ACM, SIGIR ’17, p 345–354.
[27]
Lu, Y., Dong, R., & Smyth, B. (2018). Why i like it: Multi-task learning for recommendation and explanation. In: RecSys. ACM, RecSys ’18, p 4–12.
[28]
Ma, W., Zhang, M., Cao, Y, et al. (2019). Jointly learning explainable rules for recommendation with knowledge graph. In: WWW. ACM, WWW ’19, p 1210–1221.
[29]
McAuley, J., & Leskovec, J. (2013). Hidden factors and hidden topics: Understanding rating dimensions with review text. In: RecSys, RecSys ’13, p 165–172.
[30]
Moraes, F., Yang, J., Zhang, R., et al. (2020). The role of attributes in product quality comparisons. In: CHIIR, CHIIR ’20, p 253–262.
[31]
Park YA and Gretzel U Influence of consumers’ online decision-making style on comparison shopping proneness and perceived usefulness of comparison shopping tools Journal of Electronic Commerce Research 2010 11 4 342-354
[32]
Ren, Z., Liang, S., Li, P., et al. (2017). Social collaborative viewpoint regression with explainable recommendations. In: WSDM. ACM, WSDM ’17, p 485–494
[33]
Rendle, S., & Schmidt-Thieme, L. (2010). Pairwise interaction tensor factorization for personalized tag recommendation. In: WSDM. ACM, WSDM ’10, p 81–90
[34]
Rendle, S., Freudenthaler, C., Gantner, Z., et al. (2009). BPR: Bayesian personalized ranking from implicit feedback. In: UAI, UAI’09, pp 452–461.
[35]
Tan, Y., Zhang, M., Liu, Y., et al. (2016). Rating-boosted latent topics: Understanding users and items with ratings and reviews. In: IJCAI, pp 2640–2646.
[36]
Tao, Y., Jia, Y., Wang, N., et al. (2019). The fact: Taming latent factor models for explainability with factorization trees. In: SIGIR. ACM, SIGIR’19, p 295–304.
[37]
Truong, QT., & Lauw, H. (2019). Multimodal review generation for recommender systems. In: WWW, WWW ’19, p 1864–1874,.
[38]
Vedula, N., Collins, M., Agichtein, E., et al. (2023). Generating explainable product comparisons for online shopping. In: WSDM. ACM, WSDM ’23, p 949–957.
[39]
Wang, N., Wang, H., Jia, Y., et al. (2018a). Explainable recommendation via multi-task learning in opinionated text data. In: SIGIR. ACM, SIGIR ’18, p 165–174.
[40]
Wang X, He X, Feng F, et al (2018b) Tem: Tree-enhanced embedding model for explainable recommendation. In: WWW, WWW ’18, p 1543–1552.
[41]
Wu, Y., & Ester, M. (2015). Flame: A probabilistic model combining aspect based opinion mining and collaborative filtering. In: WSDM. ACM, WSDM ’15, p 199–208.
[42]
Xian, Y., Fu, Z., Muthukrishnan, S., et al. (2019). Reinforcement knowledge graph reasoning for explainable recommendation. In: SIGIR. ACM, SIGIR’19, p 285–294.
[43]
Xie, Z., Singh, S., McAuley, J., et al. (2023). Factual and informative review generation for explainable recommendation. In: AAAI, AAAI’23.
[44]
Yang, A., Wang, N., Cai, R., et al. (2022). Comparative explanations of recommendations. In: WWW, WWW ’22, p 3113–3123,.
[45]
Zhang, Y., Lai, G., Zhang, M., et al. (2014a). Explicit factor models for explainable recommendation based on phrase-level sentiment analysis. In: SIGIR. ACM, SIGIR ’14, p 83–92,.
[46]
Zhang, Y., Zhang, H., Zhang, M., et al. (2014b). Do users rate or review? boost phrase-level sentiment labeling with review-level sentiment classification. In: SIGIR, SIGIR ’14.
[47]
Zhang, Y., Chen, X., et al. (2020). Explainable recommendation: A survey and new perspectives. Foundations and Trends® in Information Retrieval 14(1):1–101.
Index Terms
- Learning to rank aspects and opinions for comparative explanations
Index terms have been assigned to the content through auto-classification.
Recommendations
Comparative Explanations of Recommendations
WWW '22: Proceedings of the ACM Web Conference 2022As recommendation is essentially a comparative (or ranking) process, a good explanation should illustrate to users why an item is believed to be better than another, i.e., comparative explanations about the recommended items. Ideally, after reading the ...
A comparative analysis of recommender systems based on item aspect opinions extracted from user reviews
In popular applications such as e-commerce sites and social media, users provide online reviews giving personal opinions about a wide array of items, such as products, services and people. These reviews are usually in the form of free text, and ...
Learning to rank for hybrid recommendation
CIKM '12: Proceedings of the 21st ACM international conference on Information and knowledge managementMost existing recommender systems can be classified into two categories: collaborative filtering and content-based filtering. Hybrid recommender systems combine the advantages of the two for improved recommendation performance. Traditional recommender ...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
© The Author(s), under exclusive licence to Springer Science+Business Media LLC, part of Springer Nature 2025.
Publisher
Kluwer Academic Publishers
United States
Publication History
Published: 16 January 2025
Accepted: 12 December 2024
Revision received: 30 July 2024
Received: 26 May 2024
Author Tags
Qualifiers
- Research-article
Funding Sources
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 05 Mar 2025