Ai-powered Mobile Proctoring Frameworks Using Machine Learning Algorithms in Higher Education: Post-covid Trends, Challenges, and Ethical Implications
This systematic review of 20 peer-reviewed studies evaluates the post-COVID landscape of AI-powered mobile proctoring in higher education, highlighting its potential for scalable exam integrity while critically addressing significant gaps in mobile-specific research, technical reliability, and ethical concerns such as privacy and algorithmic bias.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine a university exam hall, but instead of a room full of students and a few teachers watching over them, the "hall" is the entire internet, and the students are sitting in their own living rooms, kitchens, or bedrooms. This is the reality of post-pandemic higher education.
This research paper by Bartholomew Oganda Mogoi and colleagues is like a detective's report on a new kind of "digital security guard" that universities are trying to use: AI-powered mobile proctoring.
Here is a simple breakdown of what the paper says, using everyday analogies.
1. The Problem: The "Homework Hall" is Too Big
Before the pandemic, if you wanted to give a test to 1,000 students, you needed a big hall and maybe 10 human teachers to watch them. When everyone had to learn from home, that became impossible. You can't hire 10,000 teachers to sit in 10,000 different living rooms.
Universities started using computers (PCs) with cameras to watch students. But the authors noticed a big gap: Most students in developing countries don't have powerful laptops; they have smartphones. They are like people trying to drive a race car (the PC-based system) when they only have a bicycle (the smartphone). The current systems often don't work well on phones, or they are too heavy and slow for them.
2. The Solution: The "Smartphone Security Guard"
The paper looks at a new idea: Mobile Proctoring. Instead of a heavy security guard with a clipboard, imagine a tiny, super-smart robot living inside your phone.
- How it works: This robot uses "Machine Learning" (a type of AI that learns from examples) to watch the student through the phone camera. It looks for "suspicious behavior," like looking away too much, someone else entering the room, or moving the phone strangely.
- The Goal: To make sure the exam is fair without needing a human to sit there for three hours.
3. What the Researchers Found (The "Good" and the "Bad")
The authors reviewed 20 specific studies (out of 180 they looked at) to see how this technology is doing. Here is what they found:
The Good News: It's Fast and Scalable
- The "Magic Eye": The AI is getting really good at spotting things. It can use "Computer Vision" (like a digital eye) to track where a student is looking or if a second person walks into the room.
- Lightweight: Newer versions of this AI are like "compact cars." They are small enough to run on a regular smartphone without making the phone overheat or crash, which is a huge improvement over the old, heavy systems that needed big computers.
The Bad News: It's Not Perfect (Yet)
- The "False Alarm" Problem: Sometimes the robot gets confused. If the lighting in the room is bad, or the phone camera is old, the AI might think a student is cheating when they are just reading the question. It's like a smoke detector that goes off when you just toast a piece of bread.
- The "Bias" Problem: The paper highlights a serious issue: The AI is sometimes "racist" or "unfair" by accident. Because the AI was trained mostly on photos of people with lighter skin in bright rooms, it often fails to recognize students with darker skin or those in dimly lit rooms. This means some students get flagged for cheating unfairly, just because of their environment or skin tone.
- The "Big Brother" Feeling: Students feel uncomfortable. Imagine taking a test while a robot is staring at you 24/7, recording your face and your messy bedroom. Many students feel this invades their privacy and makes them anxious, like they are being watched by a spy.
4. The Missing Piece: The "Rulebook"
The paper argues that we are building the car (the technology) faster than we are writing the traffic laws (the ethics).
- Privacy: Who owns the video of your exam? Is it deleted after the test? The paper says many schools don't have clear rules about this.
- Infrastructure: In some places, the internet is shaky. If the connection drops, the AI might think the student is cheating. The paper says we need to fix the "roads" (internet and devices) before we can drive the "cars" (AI proctoring) safely.
5. The Verdict: What's Next?
The authors conclude that AI mobile proctoring is a powerful tool, but it's currently a "work in progress."
- Don't just watch, help: Instead of just acting like a police officer catching cheaters, the technology should be designed to be fair and transparent.
- Human + Robot: The best system isn't just the robot deciding everything. It's the robot flagging a "maybe," and a human teacher making the final call. This stops the robot from making unfair mistakes.
- Build for Everyone: We need to make sure the technology works for students with old phones, bad internet, and in dark rooms, not just for those with the latest gadgets.
In a nutshell: The paper says that while AI on smartphones can help universities give fair exams to everyone, we need to fix the bugs, stop the unfair bias, and respect student privacy before we let it run the show. We need to build a system that is smart, but also kind and fair.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.