Modeling a Dialogue Strategy for Personalized Movie Recommendations
Pontus Wärnestål · Intelligent User Interfaces · 2005
This paper addresses conversational interaction in user-adaptive recommender systems. By collecting and analyzing a movie recommendation dialogue corpus, two initiative types that need to be accommodated in a conversational recommender dialogue system are identified. The initiative types are modeled in a dialogue strategy suitable for implementation. The approach is exemplified by the MADFILM movie recommender dialogue system.