Data mining por meio de análise de redes, no contexto de filtro colaborativo
Título da Revista
ISSN da Revista
Título de Volume
In this dissertation, a graphical representation of large networks based on the use of cohesion surfaces over a multidimensionally scaled thematic base is proposed as a tool for Collaborative Filtering. For its development Classic Multidimensional Scaling and Procrustes Analysis are combined in an iterative algorithm, which consolidates partial solutions into an overall continuous representation. Tested on a set of book lending transactions at the Karl A. Boedecker Library, the algorithm produces an output that is thematically interpretable and consistent, with a stress measure smaller than Classic MDS solutions. The study of representation stability in face of sampling uncertainty, based on a sampling simulation at 6 different levels of sampling probability and 500 replications for each level, provides evidence in support of algorithm results validity.