BDBComp
Parceria:
SBC
Uma Metodologia para a Caracterização de Comportamento Evolutivo em Sistemas de Recomendação

Alan CardosoWalkiria ResendeDaniel RochaFernando MourãoLeonardo RochaWagner Meira Jr.

Recommender Systems (RSs) have become increasingly importanttools for various commercial applications on the Web. Despite numerousefforts, RSs still need improvements to make recommendation moreeffective and applicable to many real scenarios. Recent studies point outthe temporal evolution as a primordial manner for improving RSs without,however, understand in detail how this evolution occurs. Thus, we proposea methodology for evolutive characterization of users and applications inorder to provide a better understanding of this temporal dynamic in RSs.Applying the methodology in a real scenario has proved to be useful even tohelp in the choice of RSs adherents of each scenario.

http://www.lbd.dcc.ufmg.br/colecoes/sbrc/2011/0049.pdf

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato: bdbcomp@lbd.dcc.ufmg.br
     Mantida por:
LBD