Detección de plagio translingüe utilizando una red semántica multilingüe

[EN] Plagiarism is defined as the unauthorized use of the original content of other authors. It is a difficult phenomenon to detect whose problem has worsened in recent years because of the Internet: a vast source of information that allows users to copy and take possession, very simply, of the orig...

Descripción completa

Detalles Bibliográficos
Autor: Franco Salvador, Marc
Tipo de recurso: tesis de maestría
Fecha de publicación:2013
País:España
Institución:Universitat Politècnica de València (UPV)
Repositorio:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Idioma:español
OAI Identifier:oai:riunet.upv.es:10251/44658
Acceso en línea:https://riunet.upv.es/handle/10251/44658
Access Level:acceso abierto
Palabra clave:Detección de plagio translingüe
Similitud textual
Red semántica multilingüe
BabelNet
Grafos de conocimiento
Cross-language plagiarism detection
Textual similarity
Multilingual semantic network
Knowledge graphs
LENGUAJES Y SISTEMAS INFORMATICOS
Máster Universitario en Inteligencia Artificial, Reconocimiento de Formas e Imagen Digital-Màster Universitari en Intel·ligència Artificial, Reconeixement de Formes i Imatge Digital
Descripción
Sumario:[EN] Plagiarism is defined as the unauthorized use of the original content of other authors. It is a difficult phenomenon to detect whose problem has worsened in recent years because of the Internet: a vast source of information that allows users to copy and take possession, very simply, of the original content of other authors work. Although plagiarism can be detected manually, given the large amount of content published, it is virtually impossible to carry out, even more if the source of plagiarism comes from documents in other languages. Currently, literature and science have strong interest in research and development of automatic monolingual and cross-language similarity detection systems, capable of detecting plagiarism among sections between documents. The Academic Community also benefits by such systems. It allows teachers to detect and discourage their students of the usual practice of copy and paste, without reference to its source, from original content obtained from Internet. In this thesis we describe the state-of-the-art in text plagiarism detection at monolingual and cross-language level. In addition, we study the use of a multilingual semantic network to create two cross-language plagiarism detection models: using a statistical dictionary, and using knowledge graphs as context models from document fragments. Experimental results are very promising. As future work, we define different research lines using knowledge graphs.