IPA - An intelligent personal assistant agent for task performance support
Gabriela Czibula, Adriana-Mihaela Guran, István Gergely Czibula, Grigoreta-Sofia Cojocar · 2009
Assisting users in performing their tasks is an important issue in human computer interaction research. A solution to deal with this challenge is to build a personal assistant agent capable to discover the user's habits, abilities, preferences, and goals, ever more accurately anticipating the user's intentions. In order to solve in an intelligent manner this problem, the assistant agent has to continuously improve its behavior based on previous experiences. By endowing the agent with the learning capability, it will become able to adapt himself to the user's behavior. This paper proposes an intelligent personal assistant agent that learns by supervision to assist users in performing specific tasks. For evaluating the performance of the agent a case study is considered, and a neural network is used by the agent to learn by supervision from its experience. We also provide a comparison of our approach with other similar existing work.