Team Formation with Relationship Strength Based on Meta Path in Heterogeneous Network

Abstract

The goal of the team formation is to discover a collaborative team which contains all the requested skills and minimizes the communication cost between members. In the previous work, the team formation problem is solved in a homogeneous social network, which would waste a lot of computation time. In this paper, we solve the team formation problem in a heterogeneous social network, it will reduce the computation time. In the team formation problem, the communication cost depends on the relationship strength between the experts. Therefore, the estimation of the relationship strength is pretty important. In the previous work, the method of calculating the relationship strength is relatively simple, and cannot well represent the strength of the relationship between experts. According to the heterogeneous network, in this paper, we present an effective method for estimate the strength of relationships between experts based on meta paths. We call it RSMP. Experiments on the DBLP dataset show the effectiveness of the proposed methods.