
[ Bristol CS | Index | ML group | Peter Flach | Papers | Presentations | Student projects ]
Student project: Autonomous Agenda Agents
Suitable For
M.Sc. Conversion (Research Project)
M.Sc. Global
4th year final project
Number of People
1
Synopsis
An Autonomous Agenda Agent (AAA) is an entity that maintains a
person's agenda. The main task of an AAA is to negotiate appointments
for meetings.
It can operate either interactively (via the web) or communicate
with other AAA's through TCP/IP.
The AAA tries to satisfy the user's preferences (group meetings together on the
same day, Wednesday = research day) as much as possible.
Such preferences are represented declaratively as rules.
In a second stage of the project these rules
should be acquired automatically by machine learning methods.
The approach is to develop plug-in modules that enhance the AAA
with learning capabilities.
Objectives
- To build a practical AAA
- To explore the use of machine learning in agents
Deliverables
An operational Autonoous Agenda Agent that can maintain a user's personal
agenda, negotiate dates for appointments respecting the user's preferences,
and possibly acquire such preferences by machine learning techniques.
Skills Required
- competence in programming (preferably Java)
- general understanding of the agent paradigm
- general understanding of machine learning
Tools/Environments
Java, WWW, TCP/IP
Benefit to Student
- To take part in some genuine and valuable artificial intelligence research
- Develop modular programming skills
Contact
Peter.Flach@cs.bris.ac.uk
Background
http://www.cs.bris.ac.uk/Research/MachineLearning/
http://www.ececs.uc.edu/~abaker/JAFMAS/
P A Flach,
Peter.Flach@bristol.ac.uk. Last modified on Thursday 17 September 1998 at 09:59. © 1998 University of Bristol