Título: Mapas Auto-Organizáveis Aplicados a Sistemas Tutores Inteligentes
Autores: Martins, Weber; Carvalho, Sirlon Diniz
Resumo: This work presents an Intelligent Tutorial System based on SOM (Self-Organizing Maps) neural networks, able to adapt and react in order to offer customized and dynamic tuition. The proposed system in implemented in web environment (and technology) to take advantages like: wide covering, portability, among others. The content, source of knowledge to be learned, has been modeled in an original way and is adequate to neural control. Two user groups have been analyzed The first one moves freely in the content, while the other group is guided by the decision of neural networks previouusly trained from the most successful free interactions. Statiscal techniques were employed to analyze the significance of sample differences to compare the two groups. Results related to the interaction time have shown significant differences in favor of the guided tutor. With relation to the acquisition of new knowledge, all users guided by the intelligent control have performed as well as the best ones whith had freedom to navigate througt th content.
Código DOI: 10.21528/CBRN2003-076
Artigo em PDF: 6CBRN_076.PDF
Arquivo BibTex: 6CBRN_076.bib