Abstract
The task of a traditional recommendation engine generally ends when it returns a list of possible answers to a user’s query, even if the user finds the answers unhelpful. Such a paradigm is incompatible with modern virtual assistants, where a user might request to filter, control, and modify results in an interactive conversational flow that is based on user context. This disclosure describes a conversational recommendation engine that provides an explanation of the recommendations provided to the user, proactively offers the user options to find better recommendations, asks personalized follow-up questions to ease the interaction, and enables the user to provide real-time feedback to update the recommendation.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Anonymous, "Interactive Recommendation Engine for Conversational Recommendations", Technical Disclosure Commons, (April 01, 2020)
https://www.tdcommons.org/dpubs_series/3091