George J. Lee's Research
I am a graduate student in the Advanced Network Architecture
group at MIT's Computer Science
and Artificial Intelligence Laboratory, or CSAIL. I am interested in
applying AI techniques such as machine learning to address
networking systems problems. Below I describe some of my previous
research. You can view a full list of my publications here.
Intelligent Network Fault Diagnosis
I am currently working on a system for intelligent network fault
diagnosis in the Knowledge Plane.
I am exploring the use of model-based diagnosis, Bayesian inference, user
agents, and distributed ontologies to deal with this challenge.
-
George Lee. CAPRI: A Common Architecture for Autonomous, Distributed Diagnosis of Internet Faults using Probabilistic Relational Models. In Proceedings of the First Workshop on Hot Topics in Autonomic Computing (HotAC I), 2006.
[PDF] [Slides (PPT)]
-
George Lee, Lindsey Poole. Diagnosis of TCP Overlay Connection Failures using Bayesian Networks. To appear in Proceedings of the SIGCOMM 2006 Workshop on Mining Network Data (MineNet-06), 2006.
[PDF]
Context-aware Event Notification for Mobile Users
I worked at NTT DoCoMo R&D
Center Network Laboratory in Japan in 2004. During that time,
we developed a context-aware event notification system for mobile
users. This system automatically learns user interests and
delivers relevant information to mobile users. In this research,
we explored the use of learning and collaborative filtering to
automatically subscribe users to topics relevant for their current
context.
-
George Lee, Takefumi Naganuma, and Shoji Kurakake. Efficient
Matching in a Context-Aware Event Notification System for Mobile
Users. (Unpublished)
[PDF] [Slides (PPT)]
-
Context-aware Collaborative Filtering for Learning Interests
of Mobile Users. Presented at CSAIL Student Seminar Series,
February 14, 2005.
[Slides (PPT)]
Automatic Network Service Selection in Dynamic Wireless Networks
In the Personal Router
project, we developed an intelligent user agent that
automatically manages connectivity for mobile users by learning
user preferences implicitly from simple, high-level input. Our
research contributions were an agent architecture for automatic
service selection and a mechanism for implicitly learning which
services users preferred in different contexts. A paper we
published at PerCom describes this system.
-
George Lee, Peyman Faratin, Steven Bauer, and John Wroclawski. A
User-Guided Cognitive Agent for Network Service Selection in
Pervasive Computing Environments. In Proceedings of Second
IEEE International Conference on Pervasive Computing and
Communications (PerCom '04), 2004.
[PDF] [Slides (PPT)]
My Master's thesis also describes the agent we developed.
-
George Lee. Automatic Service Selection in Dynamic Wireless
Networks. Master's Thesis. Massachusetts Institute of
Technology, 2003.
[PDF]
Contact me: George J. Lee
Last modified: Mon Feb 14 17:33:31 EST 2005