Computational intelligent methods for trusting in social networks

Autores: José David Núñez González
Directores de la Tesis: Manuel Graña Romay (dir. tes.)
Lectura: En la Universidad del País Vasco - Euskal Herriko Unibertsitatea ( España ) en 2016
Idioma: inglés
Tribunal Calificador de la Tesis: Michal Wozniak (presid.) , Borja Fernández Gauna (secret.) , Carlos Andrés Toro Rodríguez (voc.) , José Miguel Alonso (voc.) , Alexandre Manhaes Savio (voc.)
Enlaces
- Tesis en acceso abierto en: ADDI
Resumen
- This Thesis covers three research lines of Social Networks. The first proposed reseach line is related with Trust. Different ways of feature extraction are proposed for Trust Prediction comparing results with classic methods. The problem of bad balanced datasets is covered in this work. The second proposed reseach line is related with Recommendation Systems. Two experiments are proposed in this work. The first experiment is about recipe generation with a bread machine. The second experiment is about product generation based on rating given by users. The third research line is related with Influence Maximization. In this work a new heuristic method is proposed to give the minimal set of nodes that maximizes the influence of the network.