INSPIRED: Toward Sociable Recommendation Dialog Systems
This work addresses the problem of building more effective and human-like recommendation dialog systems for users, though it is incremental as it provides a foundational dataset and initial model improvements.
The authors tackled the lack of annotated datasets for sociable recommendation dialog systems by creating INSPIRED, a dataset of 1,001 human-human movie recommendation dialogs, and found that sociable strategies like sharing personal opinions lead to more successful recommendations, with their model incorporating these strategies outperforming baselines in evaluations.
In recommendation dialogs, humans commonly disclose their preference and make recommendations in a friendly manner. However, this is a challenge when developing a sociable recommendation dialog system, due to the lack of dialog dataset annotated with such sociable strategies. Therefore, we present INSPIRED, a new dataset of 1,001 human-human dialogs for movie recommendation with measures for successful recommendations. To better understand how humans make recommendations in communication, we design an annotation scheme related to recommendation strategies based on social science theories and annotate these dialogs. Our analysis shows that sociable recommendation strategies, such as sharing personal opinions or communicating with encouragement, more frequently lead to successful recommendations. Based on our dataset, we train end-to-end recommendation dialog systems with and without our strategy labels. In both automatic and human evaluation, our model with strategy incorporation outperforms the baseline model. This work is a first step for building sociable recommendation dialog systems with a basis of social science theories.