Explainable Recommendation via Multi-Task Learning in Opinionated Text Data

Abstract

Explaining automatically generated recommendations allows users to make more informed and accurate decisions about which results to utilize, and therefore improves their satisfaction. In this work, we develop a multi-task learning solution for explainable recommendation. Two companion learning tasks of user preference modeling for recommendation} and \textit{opinionated content modeling for explanation are integrated via a joint tensor factorization. As a result, the algorithm predicts not only a user's preference over a list of items, i.e., recommendation, but also how the user would appreciate a particular item at the feature level, i.e., opinionated textual explanation. Extensive experiments on two large collections of Amazon and Yelp reviews confirmed the effectiveness of our solution in both recommendation and explanation tasks, compared with several existing recommendation algorithms. And our extensive user study clearly demonstrates the practical value of the explainable recommendations generated by our algorithm.

10 pages, SIGIR 2018

References in corpus (2)

• BPR: Bayesian Personalized Ranking from Implicit Feedback
• Multi-Task Feature Learning Via Efficient l2,1-Norm Minimization

Cited by in corpus (37)

• CATN: Cross-Domain Recommendation for Cold-Start Users via Aspect Transfer Network
• Counterfactual Explainable Recommendation
• Explainable Outfit Recommendation with Joint Outfit Matching and Comment Generation
• PRINCE: Provider-side Interpretability with Counterfactual Explanations in Recommender Systems
• Explainability in Music Recommender Systems
• Time-aware Path Reasoning on Knowledge Graph for Recommendation
• Explainable Fairness in Recommendation
• Explaining Latent Factor Models for Recommendation with Influence Functions
• Generate Natural Language Explanations for Recommendation
• Leave No User Behind: Towards Improving the Utility of Recommender Systems for Non-mainstream Users
• From Who You Know to What You Read: Augmenting Scientific Recommendations with Implicit Social Networks
• Explanation as a Defense of Recommendation
• Counterfactual Collaborative Reasoning
• Comparative Explanations of Recommendations
• Model-agnostic vs. Model-intrinsic Interpretability for Explainable Product Search
• Impact of Tone-Aware Explanations in Recommender Systems
• The FacT: Taming Latent Factor Models for Explainability with Factorization Trees
• Model Explainability in Deep Learning Based Natural Language Processing
• Learning Post-Hoc Causal Explanations for Recommendation
• From Intrinsic to Counterfactual: On the Explainability of Contextualized Recommender Systems
• Towards Explainable Collaborative Filtering with Taste Clusters Learning
• Try This Instead: Personalized and Interpretable Substitute Recommendation
• Effects of interactivity and presentation on review-based explanations for recommendations
• A Capsule Network for Recommendation and Explaining What You Like and Dislike
• Dual Preference Distribution Learning for Item Recommendation
• Explainable Product Search with a Dynamic Relation Embedding Model
• Explainable Recommendation Systems by Generalized Additive Models with Manifest and Latent Interactions
• Explainable Recommendation with Simulated Human Feedback
• Efficient Multi-Task Learning via Generalist Recommender
• Category Aware Explainable Conversational Recommendation
• BPMR: Bayesian Probabilistic Multivariate Ranking
• Hierarchical Aspect-guided Explanation Generation for Explainable Recommendation
• Directional Multivariate Ranking
• Review Regularized Neural Collaborative Filtering
• Deep Latent Emotion Network for Multi-Task Learning
• Graphing else matters: exploiting aspect opinions and ratings in explainable graph-based recommendations
• Contrastive Learning for Aspect Representation towards Explainable Recommendation