Investigación y análisis del lenguaje subjetivo en sistemas de minería de opiniones existentes en el mercado
Fecha
2019-04
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad de Guayaquil. Facultad de Ciencias Matemáticas y Físicas. Carrera de Ingeniería En Networking y Telecomunicaciones
Resumen
Los sistemas de minerías de opiniones lidian con diversas dificultades propias del lenguaje natural humando, esto provocado por la ambigüedad del lenguaje natural y los distintos fenómenos de niveles del lenguaje, esto sucede debido que el usuario no tiene conocimiento apropiado sobre estos tipos de problemas cuando generan algún tipo de contenido u opinión en un medio de internet. El objetivo principal este proyecto de investigación es elaborar un estudio del mejor método y herramienta existente que permita clasificar en sentimiento del contenido de un texto, para así mitigar la problemática que existe al momento de la extracción de la información. La metodología empleada se basa en una investigación científica y documental para recopilar la gran mayor cantidad posible de información, así mismo, se emplea la investigación de campo por medio de encuestas para conocer el grado de conocimiento que tienen 199 estudiantes de la Carrera de Ingeniería en Networking y Telecomunicaciones dado que es una carrera tecnológica y están constantemente familiarizados con las últimas tendencias en aplicaciones y herramientas tecnológicas. Todos los objetivos propuestos en el proyecto se han cumplido de tal manera que se ha determinado que el mejor método de precisión para realizar análisis de sentimiento está basado en diccionarios léxicos y para ello se necesita de una herramienta que cumpla con todas las necesidades para implementar los diccionarios, misma denominada SentiStrength.
The systems of opinion mining deal with various difficulties inherent to human language, caused by the ambiguity of the natural language and the different phenomena of language levels, this happens because the user does not have appropriate knowledge about these types of problems when they generate some type of content or opinion in an internet medium. The main objective of this research project is to develop a study of the best existing method and tool that allows us to classify the feeling of the content of a text, in order to mitigate the problems that exist at the moment of extracting the information. The methodology used is based on a scientific and documentary research to collect the greatest possible amount of information, likewise, field research is used through surveys to know the degree of knowledge that 199 students of the Engineering Degree in Networking and Telecommunications given that it is a technological career and they are constantly familiar with the latest trends in applications and technological tools. All the objectives proposed in the project have been fulfilled in such a way that it has been determined that the best precision method to perform sentiment analysis is based on lexical dictionaries and for this a tool is needed that meets all the needs to implement the dictionaries, same called SentiStrength.
The systems of opinion mining deal with various difficulties inherent to human language, caused by the ambiguity of the natural language and the different phenomena of language levels, this happens because the user does not have appropriate knowledge about these types of problems when they generate some type of content or opinion in an internet medium. The main objective of this research project is to develop a study of the best existing method and tool that allows us to classify the feeling of the content of a text, in order to mitigate the problems that exist at the moment of extracting the information. The methodology used is based on a scientific and documentary research to collect the greatest possible amount of information, likewise, field research is used through surveys to know the degree of knowledge that 199 students of the Engineering Degree in Networking and Telecommunications given that it is a technological career and they are constantly familiar with the latest trends in applications and technological tools. All the objectives proposed in the project have been fulfilled in such a way that it has been determined that the best precision method to perform sentiment analysis is based on lexical dictionaries and for this a tool is needed that meets all the needs to implement the dictionaries, same called SentiStrength.
Descripción
PDF
Palabras clave
Minerías de opiniones, Análisis sentimientos, Lenguaje natural