Researchers at Universidad Autónoma de Madrid recently created an innovative AI-powered platform that could improve remote learning, allowing educators to securely monitor students and verify that they are attending classes or mandatory online exams.

A first prototype of this platform, called Demo-edBB, should be presented at the AAAI-23 conference on artificial intelligence in February 2022, in Washington, and a version of the article is available on the arXiv preprint server.

“Our investigative group, the BiDA-Lab at the Universidad Autónoma de Madrid, has considerable experience in biometric signals and systems, behavior analysis and AI applications, with more than 300,000 articles published in the last two decades,” he told TechXplore Roberto Daza Garcia, one of the researchers who conducted the study.

“Over the past few years, virtual education has grown tremendously, becoming the main foundation of one of the most important educational institutions and generating valuable new learning opportunities. Our group has recently worked on new technologies for e-learning, ultimately leading to the development of a platform combining biometric and behavioral analysis tools.”

EdBB, the platform created by the BiDA-Lab team, was specifically designed to improve online student assessment processes, while securing them. The platform relies on several technologies, including biometric identification tools that recognize users based on their behavior (e.g., keyboard usage patterns or “keystrokes”) or physiological data (e.g., facial recognition tools), as well as algorithms trained to detect specific behaviors (e.g. attention, stress, etc.). So far, the researchers have developed a demo version of their platform, dubbed edBB-demo, but they are currently working on the full version.

“Our platform captures different sensors from the average student’s computer (webcam, keyboard, audio, metadata, etc.) and applies different technologies in real timeto identify users, suspicious events, behavior estimation, etc., then describing them in reports for teachers,” explained Daza Garcia.

“It can capture all student sensors securely and seamlessly, while allowing students to use any other online education platform. edBB-Demo combines some of the most important advances in remote biometric and behavioral understanding of the last decade.”

The platform created by this team of researchers is based on a multimodal learning framework, a model capable of analyzing different types of data, including images, videos, audio signals and metadata. The demo version of the platform was trained on a database of learning and exam sessions, each lasting over 20 minutes, with 60 different students.

“One of the biggest concerns for educational institutions is how to prove that distance students are in fact attending an online assessment,” Daza Garcia said. “The biometric and behavioral detection technologies of the edBB platform can provide greater security in this important task, while detecting a student’s behavior, which could improve the learning process and even pave the way for new technologies to estimate students’ attention or stress levels. We are convinced that these new technologies will be fundamental in the future to offer a more personalized education to each student.”

The demo version of edBB has four key features, namely it can authenticate users with high levels of accuracy, recognize human actions in videos, estimate a student’s heart rate using webcam images and estimate a student’s attentiveness by analyzing their facial expressions. The dataset used to train the framework was recently made available online and could therefore be used to train other machine learning models.

The platform created by this team of researchers could soon help advance distance learning, enabling educators to reliably and securely verify the identity of online learners. Additionally, it could facilitate the personalization of online learning, identifying possible issues that are hindering a student’s learning, such as poor attention or high stress levels.

“We believe this is a large area that has a bright future with many challenges ahead, so now we want to continue to improve edBB-Platform“, added Daza Garcia. “We want to continue developing the lines of research we are currently working on, as well as new systems for estimating cognitive load, using multimodal facial analysis and new multimodal architectures to identify the dynamics of the student’s keyboard or mouse. In addition, we want to expand our fields of investigation in visual attention estimation, gaze tracking, response prediction, etc.”

© 2022 Science X Network