Internet Computing
and Information Science Research Group
Coordinator: Dr. Kenneth Berman
Members:
Dr.
Fred Annexstein
Dr.
Kenneth Berman
Dr.
Yizong Cheng
Dr.
Jerry Paul
Dr.
John Schlipf
Major
Research Areas:
·
Internet and Web Technologies
·
Networks and Graph Theory
·
Parallel, Distributed and Cluster Computing
·
Bioinformatics and Computational Genomics
·
Logic and Semantics
·
Human/User Centric Computing
·
Knowledge Management
·
Algorithms
·
Computer and Information Technology Applications to Education
Research
Directions and Projects:
- Information search and retrieval methods. Determining and
discovering the structure of information and the uses of information.
Analysis of structural and scalability properties of WWW, Peer-to-peer
networks, and other large-scale network structures. Network and web
crawling technologies. Methods for utilizing hyperlink digraph structures
and semantic information; Clustering and classification; ranking and
relevancy of web objects. The Internet and distributed content delivery and
retrieval applications; distributed hash tables, graph center problems,
server and facility location problems, content caching.
- Advancing web-based information search,
retrieval, and sharing methods for individuals and organizations. Drs.
Berman and Annexstein and their graduate
students have developed a new, scalable network crawler capable of quickly determining the network
state of large peer-to-peer network applications. The crawler has
successfully monitored large networks, often involving thousands of
globally distributed hosts. The crawler was designed and developed to
harness the power of parallel computing clusters, which is required in
dynamic network contexts. Using the network crawler, our research team was
one of the first to analyze a large scale P2P application known as
Gnutella. In this research we were able to determine the large scale
topological structure of this network in a highly dynamic
environment. A beta-version of the
network crawling software was released and made publicly available in
2001.
A current
project is extending the functionality of the network crawler in the following
ways to enhance the search and acquisition of knowledge: (1) Extending the
functionality of our crawler to diverse network applications, beyond the original
Gnutella topology application. (2) Developing crawling technology for diverse
Internet content applications including, crawling to analyze the specific
contents of peer-to-peer networks, the contents of public web servers, and
other accessible databases. (3) Enhancing the crawling technology to achieve a
focused network crawler, capable of searching and retrieving focused content
sources. Finally, we propose to augment our crawler technology with an
associated semantic engine. Such an engine will be programmed to act on
semantic metadata, which is required for multimedia searching applications.
- Integrating new peer-to-peer concepts in the
study of knowledge acquisition and management systems. Designing models for
trusted peer computing for the goal of developing secure sharing
information systems. Investigating reliability and reachability
in networks and networks of information with the aim of designing
algorithms and analytic models for fault-tolerant routing and searching
schemes. Applications to education
and research.
- Methods for knowledge acquisition and
management. Information contexts and application of context for knowledge
based systems. Data exploration and
search and retrieval of patterns and exception in large data repositories.
Decision support and search strategies for data objects of critical
interest. Multi-dimensional data analysis, semantic web, metadata
definition, and multimedia object searching. Applications of knowledge
management to instruction and distance learning. User-centric knowledge
systems.
- Developing computational theory and
software for inferring, analyzing, and utilizing gene networks.
Applying the result to systematic pathway analysis and treatment
discovery.
- Developing models and algorithms for message
passing systems and cluster computing.
- Designing models and analyzing reliability,
routing and security of networks and information. Designing and analyzing models for
fault-tolerant routing schemes such as acorn and directional routing
schemes. Developing a theory of
directional compasses and directional embeddings with applications to
network fault tolerance and topic-sensitive searching and information
retrieval.
- Graph visualization and drawing. Mapping out and studying topology of
peer-to-peer and small-world networks, such as Gnutella and the Internet.
Graduate
Courses:
Advanced
Algorithms 1 and 2 (20-ECES-781,782)
Internet
Algorithms (20-ECES-690)
Graph
Theory and Networks (20-ECES-842)
Mathematical Logic I, II (20-ECES-621,624)
Logic in Artificial Intelligence (20-ECES-724)
Computational Genomics (20-ECES-786)