Towards Conversational Recommendation over Multi-Type Dialogs

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 1036–1049 July 5 - 10, 2020. c⃝2020 Association for Computational Linguistics 1036 Towards Conversational Recommendation over Multi-Type Dialogs Zeming Liu1∗, Haifeng Wang2, Zheng-Yu Niu2, Hua Wu2, Wanxiang Che1†, Ting Liu1 1Research Center for Social Computing and Information Retrieval, Harbin Institute of Technology, Harbin, China 2Baidu Inc., Beijing, China {zmliu, car, tliu}@ir.hit.edu.cn {wanghaifeng, niuzhengyu, wu hua}@baidu.com Abstract We focus on the study of conversational rec- ommendation in the context of multi-type di- alogs, where the bots can proactively and naturally lead a conversation from a non- recommendation dialog (e.g., QA) to a recom- mendation dialog, taking into account user’s interests and feedback. To facilitate the study of this task, we create a human-to-human Chi- nese dialog dataset DuRecDial (about 10k di- alogs, 156k utterances), which contains mul- tiple sequential dialogs for every pair of a recommendation seeker (user) and a recom- mender (bot). In each dialog, the recom- mender proactively leads a multi-type dialog to approach recommendation targets and then makes multiple recommendations with rich in- teraction behavior. This dataset allows us to systematically investigate different parts of the overall problem, e.g., how to naturally lead a dialog, how to interact with users for recom- mendation. Finally we establish baseline re- sults on DuRecDial for future studies.1 1 Introduction In recent years, there has been a significant in- crease in the work of conversational recommen- dation due to the rise of voice-based bots (Chris- takopoulou et al., 2016; Li et al., 2018; Reschke et al., 2013; Warnestal, 2005). They focus on how to provide high-quality recommendations through dialog-based interactions with users. These work fall into two categories: (1) task-oriented dialog- modeling approaches (Christakopoulou et al., 2016; Sun and Zhang, 2018; Warnestal, 2005); (2) non- task dialog-modeling approaches with more free- form interactions (Kang et al., 2019; Li et al., 2018). ∗ This work was done at Baidu. † Corresponding author: Wanxiang Che. 1Dataset and codes are publicly available at https://github.com/PaddlePaddle/models/ tree/develop/PaddleNLP/Research/ACL2020-DuRecDial. Almost all these work focus on a single type of dialogs, either task oriented dialogs for recommen- dation, or recommendation oriented open-domain conversation. Moreover, they assume that both sides in the dialog (especially the user) are aware of the conversational goal from the beginning. In many real-world applications, there are multi- ple dialog types in human-bot conversations (called multi-type dialogs), such as chit-chat, task oriented dialogs, recommendation dialogs, and even ques- tion answering (Ram et al., 2018; Wang et al., 2014; Zhou et al., 2018b). Therefore it is crucial to study how to proactively and naturally make conversa- tional recommendation by the bots in the context of multi-type human-bot communication. For ex- ample, the bots could proactively make recommen- dations after question answering or a task dialog to improve user experience, or it could lead a dialog from chitchat to approach a given product as com- mercial advertisement. However, to our knowledge, there is less previous work on this problem. To address this challenge, we present a …

Content truncated to 500 words