• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.100.2020.tde-06022020-120251
Document
Author
Full name
José Eleandro Custódio
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2020
Supervisor
Committee
Paraboni, Ivandre (President)
Fernandez Tuesta, Esteban
Rezende, Solange Oliveira
Title in Portuguese
Atribuição autoral de textos digitais
Keywords in Portuguese
Part-of-speech. POS
Word embedding
Aprendizado de máquina
Atribuição autoral
Distorções textuais
Identificação autoral
PLN
Processamento de língua natural
Abstract in Portuguese
A atribuição autoral de textos digitais (AA) visa identificar quem é o autor de um determinado texto a partir de um conjunto de autores possíveis. Sua aplicação pode ajudar na solução de casos de escândalos de corrupção, na identificação de abusos na utilização da internet, na detecção de notícias falsas, na detecção de pseudônimos e outros. Esse trabalho apresenta um estudo que usou n-gramas de caracteres, de palavras, de anotações linguísticas (POS), modelos de representação distribuída (embeddings). Foram aplicados métodos de aprendizado de máquina e proposto um método para combinar diversos modelos. Os resultados foram avaliados nos domínios literatura, letras de músicas e mensagens de microblogs
Title in English
Authorship Attribution of digital texts
Keywords in English
Author identification
Authorship attribuition
Embedding
Machine learning
Natural language processing
NLP
Part-of-speech. POS
Text distortion
Abstract in English
Authorship attribution (AA) of digital text is a computational task which aims to identify who is the author of a text given a set of candidate authors. Its application may help to solve corruption scandals, identification of abuses on internet usage, fake news detection or pseudonyms detection. Computational methods for AA includes multivariate statistics and machine learning. This work presents a study that used n-grams of characters, words, linguistic annotations (POS) and word embeddings models. Machine learning methods were applied and it was proposed a method to combine several models. Results were evaluated in literature, song lyrics and microblogging domains
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2020-03-24
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2020. All rights reserved.