A Chemistry-Oriented Cost-Effective Expert Team Formation in Social Networks
The growth of social networks in modern information systems has enabled the collaboration of experts at an unprecedented scale. Given a social network and a task consisting of a set of required skills, Team Formation (TF) aims at finding a team of experts who can cover the required skills and can communicate in an effective manner. However, this definition has been interpreted as the problem of finding teams with minimum communication cost which neglects two aspect of team formation in real life. The first is that in reality experts are multi-skilled, hence communication cost cannot be a fixed value and should vary according to the channels employed. The second ignored aspect is disregarding teams with high expertise level who can still satisfy the required communication level. To tackle above mentioned issues, I introduce a dynamic formof communication for multi-facet relationships and use it to devise a novel approach called Chemistry Oriented Team Formation (ChemoTF) based on two new metrics; Chemistry Level and Expertise Level. Chemistry Level measures scale of communication required by the task andExpertise Level measures the overall expertise among potential teams filtered by Chemistry Level. Moreover, I adopt a personnel cost metric to filter costly teams. The experimental results on the corpus compiled for this purpose suggests that ChemoTF returns communicative and cost-effective teams with the highest expertise level compared to state-of-the-art algorithms. The corpus itself is a valuable output which contains comprehensive scholarly information in the field of computer science.