Jefferson Viana Fonseca Abreu

I'm a

About

Cybersecurity & AI.

I am a cybersecurity researcher with a deep interest in the next generation (nextG) of wireless networks. Specifically, my research focuses on cybersecurity within the context of 5G and nextG, where I explore vulnerabilities and mitigation strategies in the Open Radio Access Network (ORAN). Additionally, I specialize in the intersection of adversarial machine learning and cybersecurity. My previous studies have centered on identifying vulnerabilities in human behavior. Furthermore, I have substantial industry experience as a Systems Analyst. I am dedicated to advancing the field through rigorous academic pursuits and innovative approaches.

  • Degree: Master of Science/ PhD student
  • E-mail: jeffersonva@vt.edu
  • City: Arlington, VA, 🇺🇸

Resume

Sumary

Jefferson Viana

Cybersecurity researcher passionate about next-gen wireless networks (5G/nextG). Expertise in ORAN vulnerabilities and mitigation strategies, with focus on adversarial machine learning. Also a experienced Systems Analyst. Committed to advancing the state-of-the-art through academic rigor and innovation.

  • jeffersonva@vt.edu

Education

Doctor of Philosophy (In progress): Mobile Networks

2022 - Present

Virginia Tech, Arlington, VA 🇺🇸

In progress.

Master of Science: Informatics (Cybersecurity and AI)

2019 - 2021

University of Brasília, Brasília, BR 🇧🇷

Online social networks like Twitter provide a novel channel to allow interaction between human beings. However, its success has attracted interest in attacking and exploiting through a wide range of unethical activities, such as malicious actions to manipulate users. One of the methods to carry out these abuses is the use of bots on Twitter. Such behavior needs investigation aiming to mitigate its effects. Recently, machine learning (ML) classifiers to distinguish between real and bot accounts have proven advances. In this work, we explored the technique by constructing a multi-agent system (MAS) capable of performing the Twitter bot detection autonomously. It is based on supervised classification with three ML algorithms and a reduced set of features. When tested offline, this MAS achieved good performance with an average of AUC equal to 0.9856 and a standard deviation of 0.0199. Using it for online bot detection on a Proof of Concept results suggest that 88.19\% of bots detected were correctly labeled. On a comprehensive experiment, more than 780 thousand tweets were captured during five days. 1,597 tweets were from profiles classified as bots (678 unique profiles), indicating that the approach used is feasible and practical for the real-time bot detection problem.

Bachelor of Science: Computer Science

2013 - 2018

University of Brasília, Brasília, BR 🇧🇷

The undergraduate thesis is about the development of a mathematical-computational model to detect Twitter accounts that may be vulnerable to phishing attacks. Through descriptive, exploratory, and quantitative research based on a series of experiments, it verifies the existence of correlations between some attributes of Twitter accounts and the vulnerability of their users to phishing attacks. I executed four different incremental versions of experiments. These experiments performed fake phishing attacks directed to approximately 1287 Twitter accounts extended across 38 days. My results were analyzed through logistic regression. The regression confirmed that it's possible to create 'victim profiles' based on account attributes, making it possible to execute phishing attacks with precision better than a random choice. Moreover, it is possible to build and run a tool that executes automated social engineering attacks.

Professional Experience

Graduate Research Assistant

2022 - Present

Commonwealth Cyber Initiative (CCI), Arlington, VA 🇺🇸

  • Execution of research on cutting-edge mobile networks.

Front-end Systems Analyst

2019 - 2022

Stefanini IT Solutions, Brasília, DF 🇧🇷

  • Analysis and development of front-end systems. Outsourced for the Bank of Brazil.

Tutor

2018 - 2018

Introduction to Computer Science (University of Brasília), Brasília, DF 🇧🇷

  • Give practical classes to begginer students.

RAD Software Developer (Intern)

2016 - 2018

Federal General Accounting Office (TCU), Brasília, DF 🇧🇷

  • Development and maintenance decentralized internsystems using Oracle Apex, and support to the auditors with DB queries.

Teacher Assistantship

2015 - 2015

Introduction to information systems (University of Brasília), Brasília, DF 🇧🇷

  • Assistantship to the professor with practical classes, and correction of another students projects.

Teacher Assistantship

2014 - 2014

Data structures (University of Brasília), Brasília, DF 🇧🇷

  • Assistantship to the professor with practical classes, and correction of another students projects.

Teacher Assistantship

2013 - 2013

Basic computing (University of Brasília), Brasília, DF 🇧🇷

  • Assistantship to the professor with practical classes, and correction of another students projects.

Publications

Adaptive Beam Management for Secure mmWave Communications Using Software-Defined Radios

Adrian L Baron-Hyppolite, Jefferson Viana Fonseca Abreu, Joao F. Santos, Luiz DaSilva, Jacek Kibilda
November 2023 · 2023 IEEE Military Communications Conference · Best demo paper award 🏆

AgentBotSpotter: A Multi-agent System for Online Twitter Bot Detection

Jefferson Viana Fonseca Abreu
January 2022

A Multi-agent Approach for Online Twitter Bot Detection

Jefferson Viana Fonseca Abreu, Célia Ghedini Ralha, João José Costa Gondim
June 2021 · VI Jornadas Nacionales de Investigación en Ciberseguridad

Twitter Bot Detection with Reduced Feature Set

Jefferson Viana Fonseca Abreu, Célia Ghedini Ralha, João José Costa Gondim
May 2021 · 2020 IEEE International Conference on Intelligence and Security Informatics (ISI)

Bot Development for Social Engineering Attacks on Twitter

Jefferson Viana Fonseca Abreu, Jorge Henrique Cabral Fernandes, João José Costa Gondim, Célia Ghedini Ralha
July 2020 · arXiv preprint

Desenvolvimento interativo de um robô para execução de ataques de engenharia social no twitter 🇧🇷

Jefferson Viana Fonseca Abreu, Jorge Henrique Cabral Fernandes, João José Costa Gondim
December 2019 · Conferência Ibero-Americana WWW/Internet 2019