Aplicaciones al análisis automático del contenido provenientes de la teoría matemática de la información

Abstract

This paper analyzes the most important proposals following the Shannon and Weaver's Mathematic Theory of Communication that have influenced in proceedings of automatic content analysis. It's explained the methodological applications of this theory in our discipline, especially about information retrieval. After this, describes the mathematical models applied to automatic content analysis: Laws of Zipf and Goffman, anti-dictionaries to permuted indexes, Statistical Indexation of terms by frequencies, n-grams and stemming algorisms. Also studies the methods of relation and classification like clusters by value of discrimination and by relevance of terms: for example, methods of relations based in Graph Theory, mass core, the K-means or incremental K-means, and the ISODATA algorism. Finally, explains the scientometrics indicators as Chen's coowording and methods with learning systems