ILIA: Intelligent and Learning Interface Agent

We propose to deploy our commercial off-the-shelf (COTS) technology Learn Sesame to develop a generic Intelligent and Learning Interface Agent (ILIA) to empower application developers to design interfaces that adapt themselves to the habits and preferences of users through learning. The agent interprets sequences of events that are generated from user interactions and then produces facts that represent relationships between the user behavior patterns with associated uncertainty into a cognitive model of the user. The learned facts are then used to provide assistance in customizing interfaces to fit the user needs through non-intrusive tutoring. The proposed agent will operate as a common support application, part of the Defense Information Infrastructure (DII) Common Operating Environment (COE), and therefore be embeddable within various complex applications such as combat support, tactical, and strategic mission. Our proposed agent architecture will be fully compliant with DII COE requirements in the areas of runtime environment, style guide, architectural compatibility, and software quality.