Learning User Intentions Spanning Multiple Domains

People are able to interact with domain-specific applications in smart environments and get assistance with specific tasks. Current intelligent agents (IAs) tend to be limited to specific applications. In order to engage in more complex activities users have to directly manage a task that may span multiple applications. An ideal personal IA would be able to learn, over time, about these tasks that span different resources. This paper addresses the problem of multi-domain task assistance in the context of spoken dialog systems. We propose approaches to discover users’ high-level intentions and using this information to assist users in their task. We collected real-life smart phone usage data from 14 participants and investigated how to extract high-level intents from users’ descriptions of their activities. Our experiments show that understanding high-level tasks allows the agent to actively suggest apps relevant to pursuing particular user goals and reduce the cost of users’ self-management. Author