Towards learning resources rankings in MOOCs: A pairwise based reputation mechanism

Abstract

Reputation systems have been showed as effective mechanisms for capturing and extracting the global view a society has about some entities. Traditionally, these systems are based on capturing users' opinions through quantitative evaluations given by numerical ratings. However, it has been demonstrated that mapping opinions to numerical values might entail biased problems, skewing the reputation of some entities. In this work, we present our proposal for dealing with such problems, based on capturing opinions through comparative evaluations. Besides, we state that this mechanism can be successfully applied in MOOCs, in order to estimate the reputation of learning resources, allowing us to provide students/users with better resources.