Profile and context filtering of streaming data for a mobile personal assistant

  • Authors:
  • Alan Berfield;Jonathan Beaver;Panos K. Chrysanthis

  • Affiliations:
  • University of Pittsburgh, Pittsburgh, PA;University of Pittsburgh, Pittsburgh, PA;University of Pittsburgh, Pittsburgh, PA

  • Venue:
  • Proceedings of the 2004 ACM symposium on Applied computing
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

A key component in an ubiquitous computing environment is a personal assistant that can provide a link between devices and services and a human user. A large part of interactions in such an environment is expected to be using XML data. In this paper we present the use of profiles and context to obtain useful information from streaming documents. The user profile contains all the preferences and information on typical behavior for the user. This information can be entered by the user and learned over time. As the profile is never shared or transmitted, strict user privacy is maintained. Context information includes current day and time, and the user's location. We have implemented our ideas in an intelligent personal assistant, iPA, within the context of mobile commerce.