Key points are not available for this paper at this time.
This paper presents the results obtained from a series of experiments in automatic text categorization of MEDLINE articles. The main goal ofthis research is to build a counter propagation network and to train it in assigning MeSH phrases based on term frequency of single words from title and abstract. The experiments compare the performance of the counterpropagation network against a backpropagation neural network trained for the same purpose. Results obtained by using a set of 2,344 MEDLINE documents are presented and discussed.
Ruiz et al. (Sat,) studied this question.