Section: New Results
User Oriented Expert Finding
The goal of expert search is to find a set of people who are knowledgeable in a given topic. Experts can be demanded for the variety of purposes: problem solving, question answering, providing more detailed information on a topic, to name a few. Models for expert finding proposed so far were solely focused on returning the most knowledgeable persons as experts on a given topic   . In this work we argue that in user oriented setting the notion of the best expert depends on the user and his needs. We propose a rational user model to measure the system performance without involving a real user and show how given this model the resulted set of experts is changing relatively to the list returned by classical expert finding system. Our model is based on two factors that influence user satisfaction - time spent on the task and knowledge gained after  . Next steps in this work include experiments with real users to verify the positive impact of newly introduced user parameters and to uncover other factors that influence on user's choice.