Thesis COMPRESIÓN DE TWEETS USANDO ALGORITMO DE HUFFMAN CON MODELOS PREDEFINIDOS
Loading...
Date
2018-04
Authors
ROJAS VERA, JEREMY ISAI
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
El aumento en el uso de redes sociales ha provocado que nueva información se genere de forma masiva día a día en Internet. Este hecho causa que los textos cortos como post en redes sociales, comentarios, tweets, discusiones, etc. sean cada vez más recurrentes y porconsecuencia requieren gran capacidad para su almacenamiento.Este trabajo de memoria tiene como finalidad el estudio del comportamiento de un compresorclásico aplicado en textos pequeños, proponiendo un método para que la tasa de compresiónen estos textos disminuya.Huffman es un algoritmo de compresión clásico que posee una tabla de frecuencias de símbolosque codifica y decodifica textos. Esta tabla al anexarla a textos cortos provoca que lacompresión no sea eficiente.La generalización de estas tablas de frecuencias se denominarán modelos predefinidos decompresión, que tienen como objetivo la extracción de dichas tablas de los archivos codificadosy con ello reducir las tasas de compresión en textos diminutos.
The increase in the usage of social media networks has produced that new information startsto generate massively everyday on Internet. This fact causes that small texts like posts in socialmedia networks, posts, tweets, discussions, among others, are every time more recurrentand consequently, they need a great capacity for their storage.This work has as target the study of the behavior of a classic compressor applied in smalltexts, proposing one method in which the compression rate in this kind of texts decreases.Human is a classic compression algorithm that has a frequency table of symbols that encodesand decodes texts. This table inside of the small text produces inefficient compression.The generalization of this frequency tables will be denominated predefined models of compression,that have as objective the extraction of those tables from the encoded archives and,in that way, the decrease of the compression rate in small texts.
The increase in the usage of social media networks has produced that new information startsto generate massively everyday on Internet. This fact causes that small texts like posts in socialmedia networks, posts, tweets, discussions, among others, are every time more recurrentand consequently, they need a great capacity for their storage.This work has as target the study of the behavior of a classic compressor applied in smalltexts, proposing one method in which the compression rate in this kind of texts decreases.Human is a classic compression algorithm that has a frequency table of symbols that encodesand decodes texts. This table inside of the small text produces inefficient compression.The generalization of this frequency tables will be denominated predefined models of compression,that have as objective the extraction of those tables from the encoded archives and,in that way, the decrease of the compression rate in small texts.
Description
Catalogado desde la version PDF de la tesis.
Keywords
ALGORITMO DE HUFFMAN , COMPRESION DE DATOS (CIENCIA DE LA COMPUTACION) , REDES SOCIALES