Agents for Plan Monitoring and Repair
Abstract
A key aspect to the successful planning and execution of tasks in an agent-enabled organization is the timely access to up-to-date information. Agents making decisions on behalf of human users need to access and verify information from multiple heterogeneous information sources, such as internal organizational databases (personal schedules, staff lists), real-time sensors (traffic and weather updates), as well as many types of public information (airline schedules, restaurant listings, etc). To address the problems of accessing and verifying information from heterogeneous sources, we developed a set of techniques for learning to recognize the content of the required information. The capability makes it possible to automatically access and maintain wrappers to extract data from online sources. It also provides semantic interoperability among software agents: to verify information from other agents, identify semantic mismatches in data, as well as automatically determine the type of information being communicated by an agent.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 01, 2003
- Accession Number
- ADA413560
Entities
People
- Craig Knoblock
- Ion Muslea
- Kristina Lerman
- Steve Minton
Organizations
- University of Southern California