Notice of Online Archive

  • This page is no longer being updated and remains online for informational and historical purposes only. The information is accurate as of the last page update.

    For questions about page contents, contact the Communications Division.

Senior James Bogan (Southampton, Pa.) discovered an interest in computer security while working at Lafayette’s Network Computer Center. Now, he is looking for ways to deter computer hackers during a year-long research project.

A computer science major, Bogan is conducting a senior honors thesis examining intrusion detection in pursuit of departmental honors. He is working closely with Penny Anderson, assistant professor of computer science.

Bogan explains that two kinds of intrusion detection bring network or computer break-ins to light. Misuse detection uses pre-built “signatures” to detect an attack, while anomaly detection attempts to find deviations from normal activity.

“My thesis will focus on anomaly detection, more specifically, unsupervised anomaly detection,” says Bogan.

Unlike supervised anomaly detection, unsupervised anomaly detection works on unlabeled data. Labeling of large data, such as network connection data, can be very time-consuming, so unsupervised anomaly detection is preferred.

Bogan will use a highly technical framework to test unsupervised mining algorithms. “As of now, only a few unsupervised algorithms have been fully tested, so I felt by testing some thoroughly, I would find algorithms that work well in this area,” he says.

Bogan explains that unsupervised anomaly detection algorithms perform “data mining,” extracting trends from large amounts of data.

“Jim is planning to compare the performance of three different but related approaches to intrusion detection,” says Anderson. “The biggest challenge for him is to do the comparison in a meaningful way that eliminates spurious results.”

“This research area excites me because computer security is crucial in today’s networked environments. If I am able to make an interesting discovery in this area, it is possible that I could make an impact,” says Bogan.

Anderson notes that Bogan acquired a very good background by studying artificial intelligence with Chun Wai Liew, assistant professor of computer science, and he’s rapidly educating himself by reading research papers in the area.

“Jim has exceptional ability and, perhaps even more importantly, a lot of enthusiasm for this research,” says Anderson. “My favorite thing about Jim is the quiet, direct, and very effective way he attacks problems. Those attributes will serve him well.”

Bogan is pleased to be working with Anderson. He says she is challenging, but really helps students excel in her classes. “I have a lot of confidence in Professor Anderson’s knowledge,” he adds. “I am positive that she will be able to help me get past any roadblocks that I might run into.”

A computer science tutor, Bogan is a member of Upsilon Pi Epsilon, the computer science honor society, and belongs to the student chapter of Association of Computing Machinery. For the past two summers he has worked for BEA Systems, Inc., the world’s leading application infrastructure software company.

Categorized in: Academic News