Japanese Movie Recommendation Dialogue †

This dataset is a Japanese knowledge-grounded dialogue dataset consisting of annotated movie recommendation dialogues between humans. Every recommender's utterance is associated with movie information as external knowledge. Movie information is structured in a dictionary format that maps knowledge types (e.g., "Title", "Released Year") into corresponding knowledge contents (e.g., "Rise of Planet of the Apes", "August 5, 2011").

Download †

Please download this dataset from the following Github repository:

Github: https://github.com/ku-nlp/JMRD

Contact †

Use Github Issues

Reference †

• Takashi Kodama, Ribeka Tanaka, and Sadao Kurohashi: Construction of Hierarchical Structured Knowledge-based Recommendation Dialogue Dataset and Dialogue System, In Proceedings of The 2nd DialDoc Workshop on Document-grounded Dialogue and Conversational Question Answering, pp.83-92, Dublin, Ireland, 2022. pdf