Knowledge Enhanced Multi-Domain Recommendations in an AI Assistant Application
This work addresses the challenge of making better recommendations in conversational AI assistants by combining user interaction data from multiple domains with external knowledge, offering incremental but practical gains.
The paper tackled the problem of improving recommendations in a new domain by unifying multi-domain recommendation and knowledge graph enhancement, demonstrating significant improvement on overall recommendations and for new users across three domains.
This work explores unifying knowledge enhanced recommendation with multi-domain recommendation systems in a conversational AI assistant application. Multi-domain recommendation leverages users' interactions in previous domains to improve recommendations in a new one. Knowledge graph enhancement seeks to use external knowledge graphs to improve recommendations within a single domain. Both research threads incorporate related information to improve the recommendation task. We propose to unify these approaches: using information from interactions in other domains as well as external knowledge graphs to make predictions in a new domain that would not be possible with either information source alone. We develop a new model and demonstrate the additive benefit of these approaches on a dataset derived from millions of users' queries for content across three domains (videos, music, and books) in a live virtual assistant application. We demonstrate significant improvement on overall recommendations as well as on recommendations for new users of a domain.