Extending Eigentrust with the Max-Plus Algebra
share
Eigentrust is a simple and widely used algorithm, which quantifies trust based on the repeated application of an update matrix to a vector of initial trust values. In some cases, however, this procedure is rendered uninformative. Here, we characterise such situations and trace their origin to the algebraic conditions guaranteeing the convergence of the Power Method. We overcome the identified limitations by extending Eigentrust's core ideas into the Max-Plus Algebra. The empirical evaluation of our max-plus approach demonstrates improvements over Eigentrust.
READ FULL TEXT