Dirichlet PageRank and Ranking Algorithms Based on Trust and Distrust

Abstract

Abstract. Motivated by numerous models of representing trust and distrust within a network ranking system, we examine a quantitative vertex ranking with consideration of the influence of a subset of nodes. We propose and analyze a general ranking metric, called Dirichlet PageRank, which gives a ranking of vertices in a subset S of nodes subject to some specified conditions on the vertex boundary of S. In addition to the usual Dirichlet boundary condition (which disregards the influence of nodes outside of S), we consider general boundary conditions allowing the presence of negative (distrustful) nodes or edges. We give an efficient approximation algorithm for computing Dirichlet PageRank vectors. Furthermore, we give several algorithms for solving various trustbased ranking problems using Dirichlet PageRank with general boundary conditions.

Predictable Dynamics of Opinion Forming for Networks With Antagonistic Interactions

“…In this context, the entries of the adjacency matrix are hyperlinks, and it is known that a consistent fraction of links can in principle be classifiable as "negative" links (links from spamindexing, cloaking and other "black hat" search engine optimization practices). Several algorithms have already appeared in order to cope with them [14], [13], [42]. None of these approaches is similar to ours.…”

“…where we have applied (8) and (13). Furthermore, since H is essentially nonnegative and K ≥ 0, α ′ = e Et α ≥ 0 and β ′ = e H(t−τ ) Kβ ≥ 0, meaning that…”

Improving sybil detection in online social networks

“…• Utilization of Distrust in Social Networks [25][26][27][32][33][34][35][36] [37][38][39][40][41]. Wei et al…”

“…As mentioned in chapters 1-2, most of the existing topological Sybil detection mechanisms are designed only considering inherent trust underlying social networks, while ignore the distrust factor. However, many studies reveal that distrust plays an important role in uncovering illegitimate accounts [37,41,63,64]. For example, Yang et al [50] perform an empirical analysis of characteristics of criminal accounts' social relationships on Twitter and find that the cyber criminal ecosystem in real world is more sophisticated than anticipated, which constitutes of multi-level organizations.…”

“…The detailed and formal description as well as the insight of the unified ranking mechanism are given in the subsequent sections..To leverage trust and distrust in social networks, the unified ranking mechanism is presented based on a variant of the PageRank-like model, i.e. Personalized PageRank algorithm, which is an essential technique for ranking and prediction[37,41,63,64,66].The ranking algorithm consists of two main components. The first component is to respectively propagate benign and malicious scores from a seed set of known honest and Sybil seeds among the entire network and the second component is to integrate the trust and distrust values into a unified trustworthiness for each node, which can be used to effectively discriminate non-Sybil from Sybil nodes.…”