Mental model representation using fuzzy graphs have recently grown in popularity for decision support and knowledge representation. Finding the most important node in the model has multiple applications. This paper presents a new model for static analysis in fuzzy graphs applied to the learning process. It makes use WA operators for the aggregation of the different centrality measures. This composite measure make possible to order the nodes and select the most important in a more integral way. WA operator brings flexibility to the model. A case study to show the applicability of the proposal is presented.
Despite its usefulness and high impact there is shortcomings in knowledge based recommendation models. Among its limitations are lack of flexible models, the inclusion of linguistic information and the correct weighting of the factors involved for computing a global similarity. In this paper a new knowledge based recommendation models based on the 2-tuple linguistic representation model and OWA operators is presented. It includes data base construction, vector weights determination, client profiling, products filtering and recommendation generation. Its implementation make possible to improve reliability and interpretability in recommendations. And illustrative example is shown to demonstrate the model applicability.
Show abstract
Full Text
