{"title":"Toward joint utilization of absolute and relative bandit feedback for conversational recommendation","authors":"","doi":"10.1007/s11257-023-09388-5","DOIUrl":null,"url":null,"abstract":"<h3>Abstract</h3> <p>Conversational recommendation has been a promising solution for recent recommenders to address the cold-start problem suffered by traditional recommender systems. To actively elicit users’ dynamically changing preferences, conversational recommender systems periodically query the users’ preferences on item attributes and collect conversational feedback. However, most existing conversational recommender systems only enable users to provide one type of feedback, either absolute or relative. In practice, absolute feedback can be biased and imprecise due to users’ varying rating criteria. Relative feedback, in the meanwhile, suffers from its hardship to reveal the absolute user attitudes. Hence, asking only one type of questions throughout the whole conversation may not efficiently elicit users’ preferences of high accuracy. Moreover, many existing conversational recommender systems only allow users to provide binary feedback, which can be noisy when users do not have a particular inclination. To address the above issues, we propose a generalized conversational recommendation framework, hybrid rating-comparison conversational recommender system. The system can seamlessly ask absolute and relative questions and incorporate both types of feedback with possible neutral responses. While it is promising to utilize different types of feedback together, it can be difficult to build a joint model incorporating them as they bear different interpretations of users’ preferences. To ensure relative feedback can be effectively leveraged, we first propose a bandit algorithm, RelativeConUCB. On the basis of it, we further propose a new bandit algorithm, <span>ArcUCB</span>, to utilize jointly absolute and relative feedback with possible neutral responses for preference elicitation. The experiments on both synthetic and real-world datasets validate the advantage of our proposed methods, in comparison with existing bandit algorithms in conversational recommender systems</p>","PeriodicalId":49388,"journal":{"name":"User Modeling and User-Adapted Interaction","volume":"1 1","pages":""},"PeriodicalIF":3.0000,"publicationDate":"2024-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"User Modeling and User-Adapted Interaction","FirstCategoryId":"94","ListUrlMain":"https://doi.org/10.1007/s11257-023-09388-5","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, CYBERNETICS","Score":null,"Total":0}
引用次数: 0
Abstract
Conversational recommendation has been a promising solution for recent recommenders to address the cold-start problem suffered by traditional recommender systems. To actively elicit users’ dynamically changing preferences, conversational recommender systems periodically query the users’ preferences on item attributes and collect conversational feedback. However, most existing conversational recommender systems only enable users to provide one type of feedback, either absolute or relative. In practice, absolute feedback can be biased and imprecise due to users’ varying rating criteria. Relative feedback, in the meanwhile, suffers from its hardship to reveal the absolute user attitudes. Hence, asking only one type of questions throughout the whole conversation may not efficiently elicit users’ preferences of high accuracy. Moreover, many existing conversational recommender systems only allow users to provide binary feedback, which can be noisy when users do not have a particular inclination. To address the above issues, we propose a generalized conversational recommendation framework, hybrid rating-comparison conversational recommender system. The system can seamlessly ask absolute and relative questions and incorporate both types of feedback with possible neutral responses. While it is promising to utilize different types of feedback together, it can be difficult to build a joint model incorporating them as they bear different interpretations of users’ preferences. To ensure relative feedback can be effectively leveraged, we first propose a bandit algorithm, RelativeConUCB. On the basis of it, we further propose a new bandit algorithm, ArcUCB, to utilize jointly absolute and relative feedback with possible neutral responses for preference elicitation. The experiments on both synthetic and real-world datasets validate the advantage of our proposed methods, in comparison with existing bandit algorithms in conversational recommender systems
期刊介绍：
User Modeling and User-Adapted Interaction provides an interdisciplinary forum for the dissemination of novel and significant original research results about interactive computer systems that can adapt themselves to their users, and on the design, use, and evaluation of user models for adaptation. The journal publishes high-quality original papers from, e.g., the following areas: acquisition and formal representation of user models; conceptual models and user stereotypes for personalization; student modeling and adaptive learning; models of groups of users; user model driven personalised information discovery and retrieval; recommender systems; adaptive user interfaces and agents; adaptation for accessibility and inclusion; generic user modeling systems and tools; interoperability of user models; personalization in areas such as; affective computing; ubiquitous and mobile computing; language based interactions; multi-modal interactions; virtual and augmented reality; social media and the Web; human-robot interaction; behaviour change interventions; personalized applications in specific domains; privacy, accountability, and security of information for personalization; responsible adaptation: fairness, accountability, explainability, transparency and control; methods for the design and evaluation of user models and adaptive systems
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
