Modelling the Human Values Scale in Recommender Systems: A first approach

Abstract

The objective of this paper is two-fold. The first is to develop a methodology capable of extracting the Human Values Scale (HVS) from the user, with reference to his/her objective, subjective and emotional features, in order to improve the adaptation of user models to open environments, particularly in recommender systems. For this purpose a Coherence function is defined based on the user's emotional features profiled from his/her previous interactions with the system. This function is the mapping between the user behaviour and the relevance given by him/her in each life cycle of interactions. The second objective is to present preliminary results obtained from a case study using the objective, subjective and emotional attributes in the banking domain where the methodology was tested.