Análise e classificação de conteúdo textual.

Abstract

Since the mid-50s has been developed techniques to make as natural as possible the communication between humans and computers. One of the areas responsible for studying this type of communication is called Natural Language Processing (NLP). For this final paper, a study was made in NLP area based on authors like Stuart J. Russell, Peter Norvig, Nils Nilsson John, James F. Allen, among others. This final paper used NLP resources to analyze texts, emphasizing the lexical analysis together with statistical approaches. Based on this premise, it developed a processor of texts with the main function sort, through calculations and counts, the most important words of Portuguese language texts. This solution integrates the behavior of intelligent agent of the MaRE (Study Route Mapping) application, which is responsible for identifying content that is relevant to the purpose of a web search. Therefore, with this processor was possible to identify the statistical solution to PLN is able to successfully process the words that appear more frequently in texts. Also, the solution increased the behavior of MaRE intelligent agent on textual content rating.