Towards Conversational Recommendation over Multi-Type Dialogs
Preprint
Abstract
We propose a new task of conversational recommendation over multi-type dialogs, where the bots can proactively and naturally lead a conversation from a non-recommendation dialog (e.g., QA) to a recommendation dialog, taking into account user's interests and feedback. To facilitate the study of this task, we create a human-to-human Chinese dialog dataset \emph{DuRecDial} (about 10k dialogs, 156k utterances), which contains multiple sequential dialogs for every pair of a recommendation seeker (user) and a recommender (bot). In each dialog, the recommender proactively leads a multi-type dialog to approach recommendation targets and then makes multiple recommendations with rich interaction 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 recommendation. Finally we establish baseline results on DuRecDial for future studies. Dataset and codes are publicly available at https://github.com/PaddlePaddle/models/tree/develop/PaddleNLP/Research/ACL2020-DuRecDial.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- Knowledge Graph Grounded Goal Planning for Open-Domain Conversation GenerationProceedings of the AAAI Conference on Artificial Intelligence, 2020
- The Design and Implementation of XiaoIce, an Empathetic Social ChatbotComputational Linguistics, 2020
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding2019
- Towards Knowledge-Based Recommender Dialog System2019
- Recommendation as a Communication Game: Self-Supervised Bot-Play for Goal-oriented Dialogue2019
- A Knowledge-Grounded Neural Conversation ModelProceedings of the AAAI Conference on Artificial Intelligence, 2018
- MultiWOZ - A Large-Scale Multi-Domain Wizard-of-Oz Dataset for Task-Oriented Dialogue Modelling2018
- Towards Exploiting Background Knowledge for Building Conversation Systems2018
- Towards Implicit Content-Introducing for Generative Short-Text Conversation Systems2017
- A Diversity-Promoting Objective Function for Neural Conversation Models2016