Guilherme Henrique Messias
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Guilherme Henrique Messias has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Alan Demétrius Baria Valejo
Universidade Federal de São Carlos
8 shared publications - Sylvia Iasulaitis
Universidade Federal de São Carlos
6 shared publications - Murilo Coelho Naldi
Universidade Federal de São Carlos
3 shared publications - Tiago Gonçalves Botelho
Universidade Federal de São Carlos
3 shared publications - Eloize Rossi Marques Seno
IFSP
2 shared publications - Mariana Caravanti de Souza
Federal University of Mato Grosso do Sul
2 shared publications
Latest Publications
- 30 September 2026
- Published by Springer Nature
- p. 262-276
DensGNN: A Density Based Graph Neural Network for Node Clustering in Applications with Noise
- 30 September 2026
- Published by Springer Nature
- p. 262-276
DensGNN: A Density Based Graph Neural Network for Node Clustering in Applications with Noise
- 30 September 2026
- Published by Springer Nature
- p. 293-308
Learning from Negative Edge Quality in Graph-Based Drug–Target Interaction Prediction
- 30 September 2026
- Published by Springer Nature
- p. 293-308
Learning from Negative Edge Quality in Graph-Based Drug–Target Interaction Prediction
- 30 September 2026
- Published by Springer Nature
- p. 18-33
Contrastive Deep Graph Clustering with Dynamic Cluster Estimation
- 30 September 2026
- Published by Springer Nature
- p. 18-33
Contrastive Deep Graph Clustering with Dynamic Cluster Estimation
- 30 January 2026
- Published by Springer Nature
- p. 347-361
Deep Graph Clustering Using Graph Neural Networks and Seed Detection
- 30 January 2026
- Published by Springer Nature
- p. 347-361
Deep Graph Clustering Using Graph Neural Networks and Seed Detection
- 30 January 2026
- Published by Springer Nature
- p. 331-346
Positive and Unlabeled Graph Learning with Large Language Models for Fake News Detection
- 30 January 2026
- Published by Springer Nature
- p. 331-346
Positive and Unlabeled Graph Learning with Large Language Models for Fake News Detection
- Open Access
- 19 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 215265-215285
A Systematic Review of Graph Neural Network for Node Clustering in Homogeneous Attributed Networks
The objective of node clustering is to identify distinct groups of vertices in a graph, ensuring that similar nodes are grouped together while dissimilar ones remain apart. Building on the success of graph neural networks (GNNs) in tasks such as node classification and link prediction, several clustering strategies have been developed. Although these methods are highly effective, many share similar strengths and weaknesses, which are often difficult to discern in the absence of a clear taxonomy. Previous studies have attempted to categorize node clustering algorithms; however, most are not restricted to GNN-based methods and instead encompass a wide variety of graph learning techniques. Furthermore, despite significant progress in community detection, valuable insights from this area remain underexplored in the context of GNN-based clustering. To bridge this gap and provide a structured overview of the field, we conducted a Systematic Literature Review (SLR) on node clustering in homogeneous networks using GNNs. A total of 84 articles were reviewed based on 20 inclusion and exclusion criteria. From this analysis, we propose a comprehensive taxonomy that integrates both deep clustering and node embedding models. The findings reveal key research gaps, including the absence of transformer-based models, the high computational cost of some approaches, and dataset bias arising from the over-reliance on widely used benchmarks. The proposed taxonomy offers a clearer framework for understanding existing methodologies and guiding future research directions in GNN-based node clustering.
- 19 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 215265-215285
Open AccessA Systematic Review of Graph Neural Network for Node Clustering in Homogeneous Attributed Networks
The objective of node clustering is to identify distinct groups of vertices in a graph, ensuring that similar nodes are grouped together while dissimilar ones remain apart. Building on the success of graph neural networks (GNNs) in tasks such as node classification and link prediction, several clustering strategies have been developed. Although these methods are highly effective, many share similar strengths and weaknesses, which are often difficult to discern in the absence of a clear taxonomy. Previous studies have attempted to categorize node clustering algorithms; however, most are not restricted to GNN-based methods and instead encompass a wide variety of graph learning techniques. Furthermore, despite significant progress in community detection, valuable insights from this area remain underexplored in the context of GNN-based clustering. To bridge this gap and provide a structured overview of the field, we conducted a Systematic Literature Review (SLR) on node clustering in homogeneous networks using GNNs. A total of 84 articles were reviewed based on 20 inclusion and exclusion criteria. From this analysis, we propose a comprehensive taxonomy that integrates both deep clustering and node embedding models. The findings reveal key research gaps, including the absence of transformer-based models, the high computational cost of some approaches, and dataset bias arising from the over-reliance on widely used benchmarks. The proposed taxonomy offers a clearer framework for understanding existing methodologies and guiding future research directions in GNN-based node clustering.
- 1 August 2025
- Published by Springer Nature in Knowledge and Information Systems
- Vol. 67 (11) , 10727-10749
Graph neural networks for positive and unlabeled learning: a rewiring approach
- 1 August 2025
- Published by Springer Nature in Knowledge and Information Systems
- Vol. 67 (11) , 10727-10749
Graph neural networks for positive and unlabeled learning: a rewiring approach
- Open Access
- 20 July 2025
- Published by Data in Science Editora ltda in Latin American Data in Science
- Vol. 5 (1) , 28-37
FactPolCheckBr: a dataset of fake news fact-checked during the 2022 Brazilian presidential elections
The use of social media and instant messaging applications in contemporary society has amplified the dissemination of misleading content, reaching audiences on an unprecedented scale. Many news outlets have made considerable efforts to check the accuracy of online content, especially during election periods, which are critical moments for spreading fake news. However, the fake news verification process is labor-intensive for humans, given the volume and speed at which misinformation circulates. Numerous studies in natural language processing and machine learning have emerged in recent years seeking to investigate and develop computational models capable of detecting fake news. Algorithm training is primarily based on supervised machine learning, which relies on labeled datasets to learn the characteristic patterns of misinformation. Labeled fake news datasets in Brazilian Portuguese are scarce. This research addresses this gap developing the first fact-checked fake news dataset related to the 2022 presidential elections in Brazil, which was widely regarded as the most polarized in the country’s political history and marked by a large-scale disinformation campaign. The dataset, called FactPolCheckBr, includes 1,873 news items categorized as fake news, which were manually collected from online fact-checking platforms. The full texts of the fake news items were subsequently retrieved from the web using a scraping algorithm. Next, a clustering algorithm was applied to group similar news items, which enabled the identification of the main topics targeted by fake news during the elections. Each news item in the dataset also includes information on the candidate favored by the misinformation in that electoral context. The information was provided by political scientists who employed content analysis to examine the news texts carefully. This article presents an exploratory study of the FactPolCheckBr dataset, highlighting its key features and potential applications across various domains.
- 20 July 2025
- Published by Data in Science Editora ltda in Latin American Data in Science
- Vol. 5 (1) , 28-37
Open AccessFactPolCheckBr: a dataset of fake news fact-checked during the 2022 Brazilian presidential elections
The use of social media and instant messaging applications in contemporary society has amplified the dissemination of misleading content, reaching audiences on an unprecedented scale. Many news outlets have made considerable efforts to check the accuracy of online content, especially during election periods, which are critical moments for spreading fake news. However, the fake news verification process is labor-intensive for humans, given the volume and speed at which misinformation circulates. Numerous studies in natural language processing and machine learning have emerged in recent years seeking to investigate and develop computational models capable of detecting fake news. Algorithm training is primarily based on supervised machine learning, which relies on labeled datasets to learn the characteristic patterns of misinformation. Labeled fake news datasets in Brazilian Portuguese are scarce. This research addresses this gap developing the first fact-checked fake news dataset related to the 2022 presidential elections in Brazil, which was widely regarded as the most polarized in the country’s political history and marked by a large-scale disinformation campaign. The dataset, called FactPolCheckBr, includes 1,873 news items categorized as fake news, which were manually collected from online fact-checking platforms. The full texts of the fake news items were subsequently retrieved from the web using a scraping algorithm. Next, a clustering algorithm was applied to group similar news items, which enabled the identification of the main topics targeted by fake news during the elections. Each news item in the dataset also includes information on the candidate favored by the misinformation in that electoral context. The information was provided by political scientists who employed content analysis to examine the news texts carefully. This article presents an exploratory study of the FactPolCheckBr dataset, highlighting its key features and potential applications across various domains.