Estimation-Action-Reflection: Towards Deep Interaction Between Conversational and Recommender Systems

Abstract

Recommender systems are embracing conversational technologies to obtain user preferences dynamically, and to overcome inherent limitations of their static models. A successful Conversational Recommender System (CRS) requires proper handling of interactions between conversation and recommendation. We argue that three fundamental problems need to be solved: 1) what questions to ask regarding item attributes, 2) when to recommend items, and 3) how to adapt to the users' online feedback. To the best of our knowledge, there lacks a unified framework that addresses these problems. In this work, we fill this missing interaction framework gap by proposing a new CRS framework named Estimation-Action-Reflection, or EAR, which consists of three stages to better converse with users. (1) Estimation, which builds predictive models to estimate user preference on both items and item attributes; (2) Action, which learns a dialogue policy to determine whether to ask attributes or recommend items, based on Estimation stage and conversation history; and (3) Reflection, which updates the recommender model when a user rejects the recommendations made by the Action stage. We present two conversation scenarios on binary and enumerated questions, and conduct extensive experiments on two datasets from Yelp and LastFM, for each scenario, respectively. Our experiments demonstrate significant improvements over the state-of-the-art method CRM [32], corresponding to fewer conversation turns and a higher level of recommendation hits.

WSDM '20 oral, 9 pages, 3 figures

References in corpus (11)

• BPR: Bayesian Personalized Ranking from Implicit Feedback
• Collaborative Memory Network for Recommendation Systems
• Aspect-Aware Latent Factor Model: Rating Prediction with Ratings and Reviews
• MMALFM: Explainable Recommendation by Leveraging Reviews and Images
• Algorithms for multi-armed bandit problems
• Neural Factorization Machines for Sparse Predictive Analytics
• Towards Deep Conversational Recommendations
• Online Clustering of Bandits
• Learning Contextual Bandits in a Non-stationary Environment
• Explicit State Tracking with Semi-Supervision for Neural Dialogue Generation
• Deep Conversational Recommender in Travel

Cited by in corpus (51)

• Advances and Challenges in Conversational Recommender Systems: A Survey
• When Large Language Models Meet Personalization: Perspectives of Challenges and Opportunities
• KuaiRec: A Fully-observed Dataset and Insights for Evaluating Recommender Systems
• Large Language Models as Zero-Shot Conversational Recommenders
• Seamlessly Unifying Attributes and Items: Conversational Recommendation for Cold-Start Users
• A Survey on Reinforcement Learning for Recommender Systems
• Rethinking the Evaluation for Conversational Recommendation in the Era of Large Language Models
• Towards Question-based Recommender Systems
• Variational Reasoning over Incomplete Knowledge Graphs for Conversational Recommendation
• Recommender systems based on graph embedding techniques: A comprehensive review
• Knowledge Graphs and Pre-trained Language Models enhanced Representation Learning for Conversational Recommender Systems
• UserSimCRS: A User Simulation Toolkit for Evaluating Conversational Recommender Systems
• PiRhDy: Learning Pitch-, Rhythm-, and Dynamics-aware Embeddings for Symbolic Music
• Towards Empathetic Conversational Recommender Systems
• Towards Topic-Guided Conversational Recommender System
• Improving Conversational Recommender Systems via Knowledge Graph based Semantic Fusion
• Bundle MCR: Towards Conversational Bundle Recommendation
• Generate, Delete and Rewrite: A Three-Stage Framework for Improving Persona Consistency of Dialogue Generation
• Knowledge-Enhanced Conversational Recommendation via Transformer-based Sequential Modelling
• Interactive Interior Design Recommendation via Coarse-to-fine Multimodal Reinforcement Learning
• Hierarchical Fashion Graph Network for Personalized Outfit Recommendation
• On Interpretation and Measurement of Soft Attributes for Recommendation
• KECRS: Towards Knowledge-Enriched Conversational Recommendation System
• Towards Standard Criteria for human evaluation of Chatbots: A Survey
• Parameter-Efficient Transfer from Sequential Behaviors for User Modeling and Recommendation
• Towards Building Voice-based Conversational Recommender Systems: Datasets, Potential Solutions, and Prospects
• RecUserSim: A Realistic and Diverse User Simulator for Evaluating Conversational Recommender Systems
• RecInDial: A Unified Framework for Conversational Recommendation with Pretrained Language Models
• COOKIE: A Dataset for Conversational Recommendation over Knowledge Graphs in E-commerce
• Beyond Single Items: Exploring User Preferences in Item Sets with the Conversational Playlist Curation Dataset
• MSCRS: Multi-modal Semantic Graph Prompt Learning Framework for Conversational Recommender Systems
• RevCore: Review-augmented Conversational Recommendation
• Conversations with Search Engines: SERP-based Conversational Response Generation
• DAGFiNN: A Conversational Conference Assistant
• MATINF: A Jointly Labeled Large-Scale Dataset for Classification, Question Answering and Summarization
• Identifying Breakdowns in Conversational Recommender Systems using User Simulation
• Deep Conversational Recommender Systems: A New Frontier for Goal-Oriented Dialogue Systems
• Rethinking Conversational Recommendations: Is Decision Tree All You Need?
• Question-Aware Memory Network for Multi-hop Question Answering in Human-Robot Interaction
• A Multi-Size Neural Network with Attention Mechanism for Answer Selection
• CoSam: An Efficient Collaborative Adaptive Sampler for Recommendation
• Enhancing Factorization Machines with Generalized Metric Learning
• Knowledge Graph-enhanced Sampling for Conversational Recommender System
• BI-REC: Guided Data Analysis for Conversational Business Intelligence
• Secure Artificial Intelligence of Things for Implicit Group Recommendations
• Unified Conversational Recommendation Policy Learning via Graph-based Reinforcement Learning
• Learning Neural Templates for Recommender Dialogue System
• On Estimating the Training Cost of Conversational Recommendation Systems
• Learning to Ask Conversational Questions by Optimizing Levenshtein Distance
• Improving Conversational Recommendation with Contextual Adaptation of External Recommenders and LLM-based Reranking
• CR-Walker: Tree-Structured Graph Reasoning and Dialog Acts for Conversational Recommendation