Single-Channel EEG Dataset for Task-Based Mental Workload Analysis

Here, raw EEG data collected from 50 subjects. Three task was given each subject Tasks:
1.Relaxation by closing eyes -> workload_score=0
2.Medium math -> workload_score=0.5
3.Hard math compared to task-2 -> workload_score=1
Each EEG recording in the dataset consists of 60 seconds of continuous single-channel (FP1) data per task. A balanced dataset.

Row & Columns:
Each row for 1 second duration. All EEG data in a 1 second duration stored in neuro_raw_eeg in a row.
neuro_raw_eeg is in array json format, which can be decoded by any programming language.
There are almost 50 lakhs EEG data points, but as we stored in array of each duration, dataset dimension reduced to 8250x23.
All columns recorded from Nurosky directly except neuro_relative_ columns, its calculated by us.
is_correct column indicate that is the subject answered correctly of the given math. If correct, all the EEG data during that task assigned as is_correct=1, otherwise is_correct=0