| Popis: |
espanolEsta investigacion tiene como objetivo identificar cambios en el estilo de escritura a traves del tiempo de 7 autores de novelas de habla inglesa. Para cada autor se realizo una organizacion de las novelas de acuerdo a la fecha de publicacion. Las novelas se clasificaron en tres etapas denominadas inicial, intermedia y final; cada etapa contiene 3 novelas. Entre dos etapas consecutivas existe por lo menos 2 anos de separacion entre las fechas de publicacion de las novelas. Para resolver el problema de deteccion de cambios en el estilo de escritura a traves del tiempo se propone utilizar un enfoque basado en aprendizaje automatico supervisado. Se crearon modelos de espacio vectorial a partir de las frecuencias de uso de n-gramas de distintos tipos y longitudes. Ademas, se utilizo el algoritmo de Analisis de Componentes Principales (Principal Component Analysis, PCA) como metodo de seleccion de n-gramas. La solucion se abordo como un problema de clasificacion utilizando los algoritmos de Maquinas de Soporte Vectorial (Support Vector Machine, SVM), Naive Bayes Multinomial (Multinomial Naive Bayes, MNB), Regresion Logistica (Logistic Regression, LG) y Liblinear como clasificadores. La metrica para medir la eficiencia de los algoritmos de aprendizaje fue la exactitud (accuracy). La investigacion mostro cambios significativos en cinco de los autores con una exactitud promedio de entre 70% y 80% en los distintos tipos de n-gramas. EnglishThis research aims to identify changes in the writing style over time of 7 authors of Englishspeaking novels. For each author, an organization of the novels was carried out according to the date of publication. The novels were classified in three stages called initial, intermediate and final; each stage contains 3 novels. Between two consecutive stages there are at least 2 years of separation between the publication dates of the novels. To solve the problem of detecting changes in writing style over time, it is proposed to use a supervised automatic learning-based approach. Vector space models were created from the frequencies of use of n-grams of different types and lengths. In addition, the algorithm of Principal Component Analysis (PCA) was used as the n-gram selection method. The solution was addressed as a classification problem using the Vector Support Machine algorithms (Support Vector Machine, SVM), Naive Bayes Multinomial (Multinomial Naive Bayes, MNB), Logistic Regression (LG) and Liblinear as classifiers. The metric to measure the efficiency of the learning algorithms was accuracy. The research showed significant changes in five of the authors with an average accuracy between 70% and 80% in the different types of n-grams |
