Finding the bias and prestige of nodes in networks based on trust scores

Abstract

Many real-life graphs such as social networks and peer-topeer networks capture the relationships among the nodes by using trust scores to label the edges. Important usage of such networks includes trust prediction, finding the most reliable or trusted node in a local subgraph, etc. For many of these applications, it is crucial to assess the prestige and bias of a node. The bias of a node denotes its propensity to trust/mistrust its neighbours and is closely related to truthfulness. If a node trusts all its neighbours, its recommendation of another node as trustworthy is less reliable. It is based on the idea that the recommendation of a highly biased node should weigh less. In this paper, we propose an algorithm to compute the bias and prestige of nodes in networks where the edge weight denotes the trust score. Unlike most other graph-based algorithms, our method works even when the edge weights are not necessarily positive. The algorithm is iterative and runs in O(km) time where k is the number of iterations and m is the total number of edges in the network. The algorithm exhibits several other desirable properties. It converges to a unique value very quickly. Also, the error in bias and prestige values at any particular iteration is bounded. Further, experiments show that our model conforms well to social theories such as the balance theory (enemy of a friend is an enemy, etc.).

Research on Trust Prediction from a Sociological Perspective

“…In [9], the authors proposed an unsupervised framework incorporating low-rank matrix factorization and homophily regularization for trust prediction, and the experimental results demonstrated the effectiveness of the proposed framework. In [43], the authors proposed a method to model and compute the bias or the truthfulness of a user in trust networks. The biases of users are their propensity to trust/distrust other users.…”

A topic-biased user reputation model in rating systems

“…Recall that UR measures the degree to which a user ratings are consistent with other users' on the same items. We follow Mishra et al's approach [21] to set the variance of users' ratings as the ground truth of UR. Let u i denote the ith rater, o j denote the jth QA pair, R i j 5 represent u i 's rating to o j .…”

Assessing the Credibility of Nodes on Multiple-Relational Social Networks

“…These theories have been well evaluated in [14,15]. [16,17] compute the bias and prestige of nodes in simple networks where the edge weight denotes the trust score. These methods emphasize on single relation and neglect comprehensive utilization of multiple relations.…”

“…8 on the original network. Secondly, we compute the error removing the bias from each edge as [16]. Finally, we remove unreliability from each edge using Eq.…”

“…Table 3 shows the errors of two datasets. Here, δ o shows the error of the original graph, δ b shows the error after removing bias [16] and δ u shows the error after removing unreliability. Our method almost improves the result except in the case of "a friend's friend is a friend" where the error is a small value.…”