Students Build Machine Learning Models to Detect Cyberattacks

Alabama A&M University students are putting artificial intelligence and data analytics to the test as they develop machine learning models designed to detect cyberattacks.

AAMU is participating in the FICO Educational Analytics Challenge for the fourth consecutive year, giving students an opportunity to apply classroom knowledge to a real-world cybersecurity problem while receiving guidance from industry data scientists.

The FICO Educational Analytics Challenge is an experiential learning program that allows students to solve real-world problems using data and artificial intelligence while working with FICO data scientists. This year’s challenge, conducted in partnership with the HBCU Data Science Consortium, includes Alabama A&M University, Dillard University, Morehouse College and Fayetteville State University.

At AAMU, 18 students are divided into three teams of six for the 12-week challenge. The teams are developing machine learning models to analyze network data and determine whether activity represents a cyberattack or intrusion.

“The students are using AI to enhance cybersecurity,” said Dr. Ed Pearson, associate dean for student success and associate professor of computer science. “They will be using AI to detect cybersecurity threats on the network.”

The AAMU teams bring together students from several academic disciplines, including computer science, electrical engineering, mechanical engineering and chemistry. Pearson said students were selected through an application process that considered skills such as Python programming and statistical analysis.

Throughout the challenge, students participate in weekly lectures and scrum meetings with FICO data scientists who serve as technical advisers and mentors. Dr. Scott Zoldi, FICO’s chief analytics officer, and members of his team are working directly with the students.

At the end of the challenge in November, the three AAMU teams will present their models to FICO representatives. The models will be evaluated on factors including accuracy and latency, along with the teams’ final presentations.

FICO provides AAMU with $15,000 each year to support the challenge, Pearson said. This year, $11,100 will be awarded to the three student teams, with $6,000 for first place, $3,000 for second place and $2,100 for third place.

The remaining funds will be used to send the first-place team to the HBCU Data Science Consortium Annual Data Science Celebration, scheduled for April 9-10, 2027, at the Riverside EpiCenter in Austell, Georgia.

For senior computer science major Sudeep Joshi, the competition has become a familiar part of his AAMU experience. The Nepal native is participating for the third time.

Joshi said the challenge changes each year. Last year’s competition focused on credit card fraud detection, while this year’s challenge centers on identifying network intrusions.

“We start from the very basics of machine learning,” Joshi said. “From there, we get more introduction to the challenge and cybersecurity, and then we learn about neural networks.”

Joshi placed third during his first year in the competition and second during his second.

“This time, I want to get first,” he said.

Senior computer science major Ayomide Braimoh of Lagos, Nigeria, is participating for the first time after previously applying for the program. He said additional cybersecurity experience, including hackathons and internships, helped prepare him for this year’s challenge.

Braimoh described the project as an opportunity to build a model that can distinguish normal network activity from potential attacks.

“We are going to create our own model that can predict if this is going to be a cyberattack intrusion or if it’s not,” Braimoh said.

For junior computer science major Shreya Kadariya of Nepal, the challenge is introducing her to areas of machine learning she had not previously explored.

“It’s new for me,” Kadariya said. “I hear new terms like neural network and unsupervised learning, so I’ll have to research and learn new things. It’s been a good learning experience.”

Students participate in lectures with FICO representatives each Monday and meet with their FICO mentors in scrum sessions to ask questions, discuss progress and work through challenges. Outside those sessions, team members continue developing their models while balancing their regular coursework.

For AAMU students, the challenge provides hands-on experience using AI, machine learning, data analytics and cybersecurity skills to address the types of complex problems they may encounter in the workforce.

Dr. Ed Pearson, right, joins Alabama A&M University students participating in the 2026 FICO Analytics Challenge outside the Bond Engineering Building.