CRSLab: An Open-Source Toolkit for Building Conversational Recommender System

• DOI: 10.18653/v1/2021.acl-demo.22
• Publication Date: 2021-07-27T01:42:51Z
• Authors: Kun Zhou; Xiaolei Wang; Yuanhang Zhou; Chenzhan Shang; Yuan Cheng; Wayne Xin Zhao; Yaliang Li; Ji-Rong Wen
• Keywords: FOS: Computer and information sciences; Association (psychology); Trust-Aware Recommender Systems; Epistemology; 02 engineering and technology; Computer Science - Information Retrieval; Context-Aware Recommender Systems; Engineering; Artificial Intelligence; 0202 electrical engineering, electronic engineering, information engineering; Recommender system; Natural Language Processing; Computer Science - Computation and Language; Architectural engineering; Content-Based Recommendation; Statistical Machine Translation and Natural Language Processing; Open source; Computer science; Programming language; FOS: Philosophy, ethics and religion; World Wide Web; Philosophy; Recommender System Technologies; Joint (building); Collaborative Filtering; Computer Science; Physical Sciences; Computation and Language (cs.CL); Software; Information Retrieval (cs.IR); Information Systems

8 pages<br/>In recent years, conversational recommender system (CRS) has received much attention in the research community. However, existing studies on CRS vary in scenarios, goals and techniques, lacking unified, standardized implementation or comparison. To tackle this challenge, we propose an open-source CRS toolkit CRSLab, which provides a unified and extensible framework with highly-decoupled modules to develop CRSs. Based on this framework, we collect 6 commonly-used human-annotated CRS datasets and implement 18 models that include recent techniques such as graph neural network and pre-training models. Besides, our toolkit provides a series of automatic evaluation protocols and a human-machine interaction interface to test and compare different CRS methods. The project and documents are released at https://github.com/RUCAIBox/CRSLab.<br/>