Sistema para la calificación de documentos en base a su completitud
Loading...
Download
Official URL
Full text at PDC
Publication date
2020
Authors
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Citation
Abstract
Debido a la constante creación de documentación en la industria, aparece la necesidad de aplicar las revisiones correspondientes para evaluar el contenido de la documentación generada. Esta labor implica un enorme esfuerzo temporal, económico y humano. Por lo tanto, es de especial interés implementar un sistema que automatice este proceso, liberando así a los profesionales de esta carga de trabajo.
Por esta razón, en este trabajo se propone un sistema que tiene la capacidad de analizar el contenido de distintos documentos y de realizar estas revisiones de forma automática. El sistema propuesto, para que sea adaptable y escalable, se ha implementado de forma que puede adaptarse a distintos dominios. Su funcionamiento no se ajusta a un único tipo de documentos.
El sistema propuesto se implementa utilizando distintas técnicas de procesamiento de lenguaje natural, de extracción de información y de aprendizaje automático. En este documento se describe tanto el funcionamiento de estas técnicas como su presencia y relevancia en la industria.
Este trabajo está relacionado con un proyecto de colaboración con la empresa ECIX Group, que plantearon esta necesidad y han proporcionado todos los recursos necesarios.
Due to the constant creation of documentation in the industry, there is a need to apply the corresponding revisions to evaluate the content of the documentation generated. This work implies an enormous temporary, economic, and human effort. Therefore, it is important to implement a system that automatatizes this process, thus freeing professionals from carrying out this task. For this reason, in this project we propose a system that has the ability to analyze the content of different documents, and to realize these reviews automatically. The proposed system, to be adaptable and scalable, has been implemented so that it can be adapted to different domains. Its operation does not conform to a single type of documents. The proposed system is implemented by using different Natural Language Processing, Information Extraction, and Machine Learning techniques. This document describes how these techniques work, its presence in the industry, and its relevance. This work is related to a collaboration project with the company ECIX Group, which raised this need, and has provided all the necessary resources.
Due to the constant creation of documentation in the industry, there is a need to apply the corresponding revisions to evaluate the content of the documentation generated. This work implies an enormous temporary, economic, and human effort. Therefore, it is important to implement a system that automatatizes this process, thus freeing professionals from carrying out this task. For this reason, in this project we propose a system that has the ability to analyze the content of different documents, and to realize these reviews automatically. The proposed system, to be adaptable and scalable, has been implemented so that it can be adapted to different domains. Its operation does not conform to a single type of documents. The proposed system is implemented by using different Natural Language Processing, Information Extraction, and Machine Learning techniques. This document describes how these techniques work, its presence in the industry, and its relevance. This work is related to a collaboration project with the company ECIX Group, which raised this need, and has provided all the necessary resources.
Description
Trabajo de Fin de Máster en Ingeniería Informática, Facultad de Informática UCM, Departamento de Ingeniería de Software e Inteligencia Artificial, Curso 2019/2020