EEG Attention Monitor
Reading brainwaves to tell whether a student is paying attention.
Research into real-time student attention detection from EEG signals. The proposal, ethics application and system design are done; the pipeline uses SVM, Random Forest and EEGNet classifiers.
The question
Can a cheap EEG headset tell, in real time, whether a person is focused? EEG (electroencephalography) records the brain’s electrical activity from the scalp. It’s noisy, it drifts, and every head is different.
Where it’s at
The groundwork is done: the research proposal, the ethics application and the system design. There’s no working code yet, and I’d rather say that than pretend otherwise.
The planned pipeline compares three classifiers. SVM and Random Forest work on hand-crafted features from the signal’s frequency bands. EEGNet is a compact convolutional network built for EEG that learns from the raw signal. Comparing them shows whether the deep model earns its extra complexity on data this noisy.
The output is a live dashboard that shows attention at a glance, so a teacher doesn’t need to know what a frequency band is.
I’ll write up the build as it happens in the articles section.
