Our smartphones are quite capable, but not really very intelligent. They have more computational power than the supercomputers NASA used to reach the moon and are loaded with sensors, but lack "situational awareness" — an ability to understand their context and use that knowledge to provide better services for their users.
CSEE professor Tim Finin is one of the organizers of a workshop on Activity Context Representation: Techniques and Languages that is focused on the problems underlying making mobile devices more intelligent. The workshop will address techniques and systems to allow mobile devices model, reason about, and recognize the activities and context of their users and exploit those models to provide better services. The workshop is sponsored by the sponsored by the Association for the Advancement of Artificial Intelligence and will be held on August 7th and 8th in San Francisco as part of AAAI-11, the Twenty-Fifth Conference on Artificial Intelligence.
Finin is working with colleagues Anupam Joshi and Laura Zavala and students from the ebiquity lab on the NSF-sponsored Platys project that is working on this problem. The project is developing software for Android phones that uses machine learning algorithms to recognize their users' activities and their roles in them and selectively share this information with other devices based on user-defined privacy policies.