Publication: Generación Automática de Texto Fantástico en Español
Date
2022-09-01
Authors
Editor
Director
Advisor
Coordinator
Commentator
Reviewer
Illustrator
Access rights
info:eu-repo/semantics/openAccess
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad Nacional de Educación a Distancia (España). Escuela Técnica Superior de Ingeniería Informática. Departamento de Lenguajes y Sistemas Informáticos
Abstract
La generación automática de texto es un problema desafiante que plantea fuertes bases en las disciplinas del procesamiento del lenguaje natural e inteligencia artificial. En este trabajo analizaremos sus fortalezas, debilidades y desafíos así como su desarrollo en la actualidad. Existen multitud de enfoques para la generación automática de texto pero a día de hoy sigue siendo una tarea compleja de abordar, ya que cuenta con grandes dificultades aún por solucionar. A lo largo de estas páginas construiremos TolkienGen, un dataset en español que nos permitirá cotejar la efectividad de diferentes tipos de modelos de generación basados en redes neuronales. Evaluaremos y analizaremos los resultados con el objetivo de comparar los resultados y establecer con ellos un análisis de la situación actual.
Automatic text generation is a challenging problem that has strong foundations in the disciplines of natural language processing and artificial intelligence. In this work we will analyze its strengths, weaknesses and challenges, as well as its current status. There are many approaches to automatic text generation, but today it is still a complex task to approach, since it has great difficulties yet to be solved. Throughout these pages we will build TolkienGen, a dataset in spanish that will allows us to compare the effectiveness of different types of generation models based on neural networks. We will evaluate and analyze the results in order to compare the results and establish an analysis of the current situation.
Automatic text generation is a challenging problem that has strong foundations in the disciplines of natural language processing and artificial intelligence. In this work we will analyze its strengths, weaknesses and challenges, as well as its current status. There are many approaches to automatic text generation, but today it is still a complex task to approach, since it has great difficulties yet to be solved. Throughout these pages we will build TolkienGen, a dataset in spanish that will allows us to compare the effectiveness of different types of generation models based on neural networks. We will evaluate and analyze the results in order to compare the results and establish an analysis of the current situation.
Description
UNESCO Categories
Keywords
Citation
Center
E.T.S. de Ingeniería Informática
Department
Lenguajes y Sistemas Informáticos