Web recommendation systems using agents

• Autor(a) principal: Morais, A. Jorge
• Data de Publicação: 2018
• Outros Autores: Neto, Joaquim, Oliveira, Eugénio, Jorge, Alípio Mário
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: https://doi.org/10.34627/rcc.v8i0.94

The growth of the Web has brought several problems for users. Today the vast amount of information on some web sites makes useful information finding very difficult. The objectives of the owners of the web sites and users do not always coincide. The knowledge of patterns of user visits is crucial to the owners to transform and adapt their web site. This is the adaptive website principle: the website adapts to improve the user experience. Some algorithms have been proposed to tailor a website. In this paper, we describe a proposal for a web recommendation system based on agents that combines two algorithms: association rules and collaborative filtering. Both algorithms are incremental and work with binary data. The results show that, in some situations, the multi-agent approach overcomes the predictive capacity of individual agents.