Community-based collaboration recommendation to support mixed decision-making support

With the spread of organizations and especially distributed ones, teams are highly dynamic and people are continually seeking new collaborators. In this context, we propose the idea of mixed decision support systems where individuals seek potential collaborators to navigate between individual and collective problem solving. In order to support mixed decision-making within organizations, we propose an approach of collaboration recommendation that relies on a community detection technique to find potential collaborators that can help in problem solving. Our contribution consists of combining two community detection approaches (modularity optimization and classical approaches) into one approach by proposing a combined metric that considers both social connections and homophily of participants. Then, we rely on a computational optimization technique (i.e. Particle Swarm Optimization) to maximize this combined quality. Finally, a case study illustrates practical uses of the proposed solution and a comparison with another community detection approach evaluates its performance.