Impact of biased edge weights and generalized modularity for positively weighted complete networks

Abstract

The objectivity of judges in a competition where determine the ranking of the contestants with judges grade are prone to be questioned, undermining the public’s trust in the fairness of the competition. In this study, we showed how we could identify biased scores and measure their effect on our interpretation of the network structure by modeling the judge-contestant competition as a network. Analyzed a recent well-publicized scoring controversy competition as an example, we showed that a single uncharacteristic score could distort our inference of the community structure. The results suggest that the importance of detecting and eliminating biases in a network.