Um modelo algébrico para representação, indexação e classificação automática de documentos digitais.

摘要

This paper introduce the idea of representing, indexing and automatically classifying digitaldocuments. The vectorial model of representing documents is simple and allows us to deal with the classification of a great amount of digital documents which were loaded daily in almost 35 Brazilian Digital Library of Thesis and Dissertation. We expect to have another 20 libraries by the end of this year. Using a sample of real documents, we compare this methodology of classification to that done byspecialists. The results show that this methodology is promising in reducing the effort of specialistswhen performing such task.