• 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.55.2023.tde-03012024-174804
Document
Author
Full name
Bruno Fernandes Bessa de Oliveira
Institute/School/College
Knowledge Area
Date of Defense
Published
São Carlos, 2023
Supervisor
Committee
Lopes, Alneu de Andrade (President)
Faleiros, Thiago de Paulo
Liang, Zhao
Rodrigues, Francisco Aparecido
Title in English
An application of graph neural networks on topic modelling with bi-partite graphs
Keywords in English
Graph deep learning
Machine learning
Neural networks
Topic modelling
Abstract in English
Graphs are data structures proper to represent real-world objects and their relationships having been widely studied in theory and with multiple examples of applications in industries and academic research. Applying graph-based data in machine learning had a significant advance with the proposal of Graph Neural Networks (GNNs), allowing the representation of this type of data in algorithms that can retain features from the graph without the need for preprocessing stage. This master's dissertation presents an analysis of GNNs and proposes an application on text classification using topic modelling to create descriptive variables in bi-partite graphs.
Title in Portuguese
Uma aplicação de redes neurais de grafos em modelagem de tópicos com grafos bipartidos
Keywords in Portuguese
Aprendizado de máquina
Deep learning com grafos
Modelagem de tópicos
Redes neurais
Abstract in Portuguese
Grafos são estruturas de dados adequadas para representar objetos do mundo real e suas inter-relações, tendo sido amplamente estudados teoricamente e com múltiplos exemplos de aplicações na indústria e pesquisa acadêmica. A aplicação de dados originados de grafos em aprendizados de máquina teve um significante avanço com a proposta das Redes Neurais de Grafos (Graph Neural Networks, ou GNNs), permitindo a representação deste tipo de dados em algoritmos que são capazes de preservar as características do grafo sem necessidade de pré-processamento. Nesta dissertação apresentamos uma análise das redes neurais de grafos e uma proposta de aplicação no contexto de classificação de textos utilizando modelagem e tópicos para criação de variáveis descritivas em grafos bipartidos.
 
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
2024-01-04
 
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 © 2001-2024. All rights reserved.