In high school in Greece, Georgia Kastidou had set her heart on the very competitive aeronautical engineering program at the country's air force university, which admitted just five women per year. But there were several other spaces on the application form - where else should she apply? On a whim, she added computer science as a secondary choice on her application, and got in to the program at the University of Ioannina.
She thought she'd try it for a year.
At the beginning it was tough. I didn't know what the Internet was. We're talking 1998. You guys here were doing on-line shopping!
Yet she rose to the challenge and pretty soon caught up to students who were very knowledgeable in computer science before they joined the department.
Eventually, Kastidou chose to come to Waterloo to do her doctoral studies. She began in the computer networks group working with Jay Black but developed an interest in artificial intelligence through the CS785 Intelligent Computer Interfaces course and switched to become a student of Robin Cohen. Under Cohen's supervision, she studied Trust & Reputation Modeling, Multi-agent Systems, Incentive Mechanisms, and Mechanism Design.
Kastidou's research is applicable to e-commerce and indeed to any on-line community where members must assess the trustworthiness of those they are interacting with. She is incorporating ideas from economics to develop electronic mechanisms that will act as incentives for participants to act well, and disincentives to malicious behaviour in multi-agent systems. She is also interested in the issue of reputation portability: how do you bring your 'good name' with you from one on-line community to another? By the same token, if word of a user's bad conduct gets around, will this make online communities behave better?
Kastidou loves computer science, and wants others to understand how crucial it is to so many facets of our lives.
Some think that computer science makes no social contribution. They think it's about hackers and nerds playing games all night. But that's not true. In a disaster, for example, people need help fast. They need supplies, doctors, transportation. Imagine how long that would take to coordinate without computers, how many people would die without them! I want people to understand how important this area is to their own lives, and then to make their own choice about it.