| Resumen |
This paper is intended to study the existing classification and information retrieval techniques in order to use an algorithm that will group the a set of documents. Therefore, the unfolding of knowledge in texts is selected as the proper methodology to be followed and the steps are explained in order to reach the unsupervised documents classification. After conducting an experiment with three of the most known methods of unsupervised documents classification and the assessment of the results with the Silhoutte index, it could be observed that the better grouping was with four groups, whose main characteristic was to deal with subjects such as: information management information, systems management, artificial intelligence, and digital image processing. |
