Glauber Dias Gonçalves
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Glauber Dias Gonçalves has a total of 63 co-authors. Here are the top co-authors by number of shared publications:
- Francisco Airton Silva
Federal University of Piauí (UFPI), Picos, Brazil
9 shared publications - Alex Borges Vieira
Computer Science Department, Universidade Federal de Juiz de Fora, Juiz de Fora, Brazil
6 shared publications - André Soares
Universidade Federal do Piauí
5 shared publications - Saulo Moraes Villela
Federal University of Juiz de Fora
5 shared publications - Heder Soares Bernardino
Universidade Federal de Juiz de Fora (UFJF)
4 shared publications - Carlos Melo
Universidade Federal do Piauí
3 shared publications
Latest Publications
- Open Access
- 12 October 2026
- Published by Elsevier in SSRN Electronic Journal
Event-Level Stance Analysis of Public Reactions with Lightweight Event Detection
Urban violence events are increasingly documented and debated through multimedia content shared on online platforms, where videos and their associated textual reactions provide a rich record of both the events themselves and how the public responds to them. Existing computational approaches typically address event identification and opinion analysis separately, while few provide a scalable mechanism for linking automatically discovered events to the distributions of public reactions they generate. This paper proposes a platform-agnostic, event-aware analytical framework that combines lightweight event clustering with stance detection in Brazilian Portuguese. Video metadata and associated comments are transformed into structured analytical units through a pipeline comprising data collection, semantic enrichment, hybrid annotation, supervised stance classification, and unsupervised clustering. We investigate three research questions: (i) the extent to which lightweight textual and temporal signals can recover real-world events from noisy multimedia streams; (ii) the robustness of Brazilian Portuguese stance classification when applied to naturally occurring online reactions; and (iii) the degree to which aggregating stances at the automatically detected event level reveals patterns hidden by global monitoring. Event identification relies on a single redesigned Temporal Heuristic compared against a GPT-4 baseline; prior evidence that a competing semantic heuristic and automatic transcripts provide limited additional value in this domain motivated this narrower scope (Rocha et al., 2025). The framework is evaluated through a YouTube-based case study comprising more than 1,400 manually labelled videos for event clustering and over one million monitored comments for event-level aggregation. The results show that the Temporal Heuristic reaches 0.94 accuracy in the high-repercussion scenario and 0.85 accuracy in the primary all-events benchmark, substantially outperforming the GPT-4 baseline in every scenario. For stance detection, a fine-tuned BERTimbau model obtains per-class F1-scores between 0.88 and 0.91 on the YouTube dataset. Event-level aggregation reveals substantial heterogeneity in public reactions across events. These findings demonstrate the value of linking automatically detected events to stance distributions as a scalable approach for studying public reactions to real-world events represented in online multimedia content.
- 12 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessEvent-Level Stance Analysis of Public Reactions with Lightweight Event Detection
Urban violence events are increasingly documented and debated through multimedia content shared on online platforms, where videos and their associated textual reactions provide a rich record of both the events themselves and how the public responds to them. Existing computational approaches typically address event identification and opinion analysis separately, while few provide a scalable mechanism for linking automatically discovered events to the distributions of public reactions they generate. This paper proposes a platform-agnostic, event-aware analytical framework that combines lightweight event clustering with stance detection in Brazilian Portuguese. Video metadata and associated comments are transformed into structured analytical units through a pipeline comprising data collection, semantic enrichment, hybrid annotation, supervised stance classification, and unsupervised clustering. We investigate three research questions: (i) the extent to which lightweight textual and temporal signals can recover real-world events from noisy multimedia streams; (ii) the robustness of Brazilian Portuguese stance classification when applied to naturally occurring online reactions; and (iii) the degree to which aggregating stances at the automatically detected event level reveals patterns hidden by global monitoring. Event identification relies on a single redesigned Temporal Heuristic compared against a GPT-4 baseline; prior evidence that a competing semantic heuristic and automatic transcripts provide limited additional value in this domain motivated this narrower scope (Rocha et al., 2025). The framework is evaluated through a YouTube-based case study comprising more than 1,400 manually labelled videos for event clustering and over one million monitored comments for event-level aggregation. The results show that the Temporal Heuristic reaches 0.94 accuracy in the high-repercussion scenario and 0.85 accuracy in the primary all-events benchmark, substantially outperforming the GPT-4 baseline in every scenario. For stance detection, a fine-tuned BERTimbau model obtains per-class F1-scores between 0.88 and 0.91 on the YouTube dataset. Event-level aggregation reveals substantial heterogeneity in public reactions across events. These findings demonstrate the value of linking automatically detected events to stance distributions as a scalable approach for studying public reactions to real-world events represented in online multimedia content.
- Open Access
- 25 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 58-63
Socio-Technical Constraints and Cross-Platform Model Generalization: Text Classification in Brazilian Public Security
- 25 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 58-63
Open AccessSocio-Technical Constraints and Cross-Platform Model Generalization: Text Classification in Brazilian Public Security
- 30 January 2026
- Published by Springer Nature
- p. 242-255
NFT Collection Classification: A Multimodal Approach Integrating Metadata and Textual Embeddings with Supervised Learning
- 30 January 2026
- Published by Springer Nature
- p. 242-255
NFT Collection Classification: A Multimodal Approach Integrating Metadata and Textual Embeddings with Supervised Learning
- 2 January 2026
- Published by Springer Nature
- p. 283-295
A Blockchain Review Based on the Scalability Trilemma
- 2 January 2026
- Published by Springer Nature
- p. 283-295
A Blockchain Review Based on the Scalability Trilemma
- 19 December 2025
- Published by Elsevier in Future Generation Computer Systems
- Vol. 179
A stochastic performance model for evaluating ethereum layer-2 rollups
- 19 December 2025
- Published by Elsevier in Future Generation Computer Systems
- Vol. 179
A stochastic performance model for evaluating ethereum layer-2 rollups
- Open Access
- 10 November 2025
- Published by Sociedade Brasileira de Computação
- p. 301-305
Decifrando o Mercado de NFTs: A Influência das Conexões de Rede e Transações no Valor de Ativos
The market for Non-Fungible Tokens (NFTs) continues to evolve, yet it still lacks robust methodologies to estimate the future value of its assets. Unlike traditional financial markets, NFT pricing is challenged by intangible factors such as the artistic nature of the items and the influence of social and transactional networks among buyers and sellers. This study investigates whether the structural position of participants in the transaction network can serve as a relevant predictor of the future value of NFTs. To this end, we reconstructed the NFT trading network for the period 2020–2021, extracted both structural and transactional metrics of the participants, and applied supervised machine learning models, including deep neural networks. The results demonstrate the feasibility of the proposed approach, achieving 74% accuracy and a global F1-Score of 72%. Interpretability analysis using SHAP values revealed that, in addition to historical price averages, network metrics such as degree and neighborhood significantly contribute to prediction. These findings highlight the role of network dynamics in NFT valuation and point toward promising directions for more transparent and evidence-based pricing methodologies.
- 10 November 2025
- Published by Sociedade Brasileira de Computação
- p. 301-305
Open AccessDecifrando o Mercado de NFTs: A Influência das Conexões de Rede e Transações no Valor de Ativos
The market for Non-Fungible Tokens (NFTs) continues to evolve, yet it still lacks robust methodologies to estimate the future value of its assets. Unlike traditional financial markets, NFT pricing is challenged by intangible factors such as the artistic nature of the items and the influence of social and transactional networks among buyers and sellers. This study investigates whether the structural position of participants in the transaction network can serve as a relevant predictor of the future value of NFTs. To this end, we reconstructed the NFT trading network for the period 2020–2021, extracted both structural and transactional metrics of the participants, and applied supervised machine learning models, including deep neural networks. The results demonstrate the feasibility of the proposed approach, achieving 74% accuracy and a global F1-Score of 72%. Interpretability analysis using SHAP values revealed that, in addition to historical price averages, network metrics such as degree and neighborhood significantly contribute to prediction. These findings highlight the role of network dynamics in NFT valuation and point toward promising directions for more transparent and evidence-based pricing methodologies.
- Open Access
- 18 August 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Internet Technology
- Vol. 25 (3) , 1-22
Analysis of the Behavior of Ethereum Accounts During an Economic Impact Event
One of the main events involving the world economy in 2022 was the beginning of the war between Russia and Ukraine. This event offers an opportunity to analyze how a large-magnitude world event can affect the use of cryptocurrencies. Ethereum is one of the most prominent and widely used cryptocurrency platforms and, as such, provides a valuable case study for this scenario. This work investigates the behavior of accounts and their transactions on the Ethereum network during this event. For this purpose, we collect all Ethereum transactions during two distinct periods: (i) during the month the conflict began, and (ii) during the previous year. We organized a dataset with the accounts involved in these transactions and the subset of these accounts that interacted with a service within Ethereum named Flashbots Auction. Flashbots Auction is crucial as it addresses issues regarding transaction ordering and miners exploiting that ordering to make profit. Then, we model temporal graphs in which each vertex represents an account, and each edge represents a transaction between two accounts. We analyzed the behavior of these accounts via graph metrics for both groups during each observed time window. The results show changes in account behavior and activity, as well as variations in daily transaction volume.
- 18 August 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Internet Technology
- Vol. 25 (3) , 1-22
Open AccessAnalysis of the Behavior of Ethereum Accounts During an Economic Impact Event
One of the main events involving the world economy in 2022 was the beginning of the war between Russia and Ukraine. This event offers an opportunity to analyze how a large-magnitude world event can affect the use of cryptocurrencies. Ethereum is one of the most prominent and widely used cryptocurrency platforms and, as such, provides a valuable case study for this scenario. This work investigates the behavior of accounts and their transactions on the Ethereum network during this event. For this purpose, we collect all Ethereum transactions during two distinct periods: (i) during the month the conflict began, and (ii) during the previous year. We organized a dataset with the accounts involved in these transactions and the subset of these accounts that interacted with a service within Ethereum named Flashbots Auction. Flashbots Auction is crucial as it addresses issues regarding transaction ordering and miners exploiting that ordering to make profit. Then, we model temporal graphs in which each vertex represents an account, and each edge represents a transaction between two accounts. We analyzed the behavior of these accounts via graph metrics for both groups during each observed time window. The results show changes in account behavior and activity, as well as variations in daily transaction volume.
- Open Access
- 19 May 2025
- Published by Sociedade Brasileira de Computação
- p. 602-615
ContAudIT: Uma Proposta Fim-a-Fim para Auditoria Contínua de Gerência de Mudanças em TI usando Blockchain
Mudanças em TI são essenciais para a operação contínua de organizações modernas, mas o gerenciamento inadequado dessas mudanças pode representar riscos significativos para a continuidade dos negócios. Nesse contexto, auditorias desempenham um papel fundamental em garantir a gerência de mudanças segundo procedimentos homologados. Tradicionalmente, firmas de auditoria realizam inspeções periódicas dos sistemas de TI e registros de mudanças. No entanto, o aumento do volume, complexidade e automação das mudanças torna desafiador assegurar a conformidade entre as inspeções de auditoria. Este artigo propõe o ContAudIT, uma abordagem baseada em blockchain para auditoria contínua de mudanças em TI. A solução integra um framework de orquestração de mudanças com um registro em blockchain que certifica e registra cada mudança implementada. A cadeia de mudanças entre eventos de inspeções permite que auditores verifiquem que apenas mudanças certificadas foram realizadas na infraestrutura.
- 19 May 2025
- Published by Sociedade Brasileira de Computação
- p. 602-615
Open AccessContAudIT: Uma Proposta Fim-a-Fim para Auditoria Contínua de Gerência de Mudanças em TI usando Blockchain
Mudanças em TI são essenciais para a operação contínua de organizações modernas, mas o gerenciamento inadequado dessas mudanças pode representar riscos significativos para a continuidade dos negócios. Nesse contexto, auditorias desempenham um papel fundamental em garantir a gerência de mudanças segundo procedimentos homologados. Tradicionalmente, firmas de auditoria realizam inspeções periódicas dos sistemas de TI e registros de mudanças. No entanto, o aumento do volume, complexidade e automação das mudanças torna desafiador assegurar a conformidade entre as inspeções de auditoria. Este artigo propõe o ContAudIT, uma abordagem baseada em blockchain para auditoria contínua de mudanças em TI. A solução integra um framework de orquestração de mudanças com um registro em blockchain que certifica e registra cada mudança implementada. A cadeia de mudanças entre eventos de inspeções permite que auditores verifiquem que apenas mudanças certificadas foram realizadas na infraestrutura.
- Open Access
A Comprehensive Hyperledger Fabric Performance Evaluation based on Resources Capacity Planning
Hyperledger Fabric is a platform for permissioned blockchain networks that enables secure and auditable distributed data storage for enterprise applications. There is a growing interest in applications based on this platform, but its use requires the configuration of different blockchain parameters. Various configurations impact the system's non-functional qualities, especially performance and cost. In this article, we propose a Stochastic Petri Net to model the performance of the Hyperledger Fabric platform with different blockchain parameters, computer capacity, and transaction rates. We also present a set of case studies to demonstrate the feasibility of the proposed model. This model serves as a practical guide to help administrators of permissioned blockchain networks find the best performance for their applications. The proposed model allowed us to identify the block size that leads to a high mean response time (ranging from 1 to 25 seconds) caused by a change in the arrival rate.
Open AccessA Comprehensive Hyperledger Fabric Performance Evaluation based on Resources Capacity Planning
Hyperledger Fabric is a platform for permissioned blockchain networks that enables secure and auditable distributed data storage for enterprise applications. There is a growing interest in applications based on this platform, but its use requires the configuration of different blockchain parameters. Various configurations impact the system's non-functional qualities, especially performance and cost. In this article, we propose a Stochastic Petri Net to model the performance of the Hyperledger Fabric platform with different blockchain parameters, computer capacity, and transaction rates. We also present a set of case studies to demonstrate the feasibility of the proposed model. This model serves as a practical guide to help administrators of permissioned blockchain networks find the best performance for their applications. The proposed model allowed us to identify the block size that leads to a high mean response time (ranging from 1 to 25 seconds) caused by a change in the arrival rate.
- Open Access
Transactional Dynamics in Hyperledger Fabric: A Stochastic Modeling and Performance Evaluation of Permissioned Blockchains
Blockchain, often integrated with distributed systems and security enhancements, has significant potential in various industries. However, environmental concerns and the efficiency of consortia-controlled permissioned networks remain critical issues. We use a Stochastic Petri Net model to analyze transaction flows in Hyperledger Fabric networks, achieving a 95% confidence interval for response times. This model enables administrators to assess the impact of system changes on resource utilization. Sensitivity analysis reveals major factors influencing response times and throughput. Our case studies demonstrate that block size can alter throughput and response times by up to 200%, underscoring the need for performance optimization with resource efficiency.
Open AccessTransactional Dynamics in Hyperledger Fabric: A Stochastic Modeling and Performance Evaluation of Permissioned Blockchains
Blockchain, often integrated with distributed systems and security enhancements, has significant potential in various industries. However, environmental concerns and the efficiency of consortia-controlled permissioned networks remain critical issues. We use a Stochastic Petri Net model to analyze transaction flows in Hyperledger Fabric networks, achieving a 95% confidence interval for response times. This model enables administrators to assess the impact of system changes on resource utilization. Sensitivity analysis reveals major factors influencing response times and throughput. Our case studies demonstrate that block size can alter throughput and response times by up to 200%, underscoring the need for performance optimization with resource efficiency.