Expert selection in recommendation networks

The Weighted Top Candidate (WTC) algorithm is an expert identification method that presents an alternative for network centralities. Its main advantage is its axiomatic characterization that shows why it is especially suitable for a number of applications. The WTC algorithm, upon receiving a recommendation network, produces a list of agents that can be deemed experts of the field. With a parameter we can adjust how exclusive our list should be. By completely relaxing the parameter, we obtain the largest stable set - agents that can qualify as experts under the mildest conditions. With a strict setup, we obtain a short list of the absolute elite.

 

We present three case studies to show the method's versatility. First, we demonstrate the algorithm on a citation database compiled from game theoretic literature published between 2008–2017.  We compile a ranking of the best game theoretical research institutions and compare it to rankings based on other network centralities. 

 

In our second study we demonstrate how powerful WTC is identifying innovators and early adopters on social networks. We examine two real-life networks (iWiW and Pokec) and calculate network centralities for each node. We compare the average registration date of the top 1000 agents for each network centrality. WTC significantly outperforms every other tested method.

 

Finally, we look at the members of the ITRE Committee of the European Parliament. We use the amendment cosponsorship graph as a recommendation network and determine the experts in each party group.