Dark Mode
EMD PPG Stress Feature Set
Data Science and Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This resource presents calculated features derived from Photoplethysmogram (PPG) signals after the application of Empirical Mode Decomposition (EMD) methodology. It was created through a feature engineering process designed to facilitate the estimation and classification of mental workload or stress levels. The resultant data is structured for use in machine learning models, particularly for binary classification tasks related to physiological stress analysis.
Columns
The data structure includes a set counter to track the sequence of PPG measurements and five statistical features extracted specifically from the first Intrinsic Mode Function (IMF 1) generated by the EMD process, alongside a binary target variable. The primary file details six columns:
- Set_no: The serial identifier or position of the original PPG data set.
- Imf_1_MEAN: The calculated mean value of the first IMF for the corresponding set.
- Imf_1_MIN: The calculated minimum value of the first IMF for the corresponding set.
- Imf_1_MAX: The calculated maximum value of the first IMF for the corresponding set.
- Imf_1_SKEWNESS: The measure of skewness of the first IMF for the corresponding set.
- Label: The target class indicating the mental workload level (a binary value of 0 or 1).
Distribution
The dataset is available in two distinct CSV files:
emd_1_imfs.csv and emd_2_imfs.csv. Both files contain the set counter, statistical features relating to the number of IMFs generated, and the corresponding stress level label. The sample data provided details 8,800 records across the six columns, with no missing values. The data is structured as extracted features, ready for model input. Updates are expected quarterly.Usage
This data is ideal for research in bio-signal processing and applied machine learning. It can be used to build and train binary classifier models, such as Support Vector Machine (SVM) models, aimed at accurately estimating stress levels from physiological signals. Suitable applications include developing predictive models for mental workload, time series analysis of bio-features, and advanced feature engineering study.
Coverage
The sources indicate the data originates from feature engineering on PPG signals intended for stress classification. Specific geographic, demographic, or precise time range coverage details regarding the original physiological data source are not specified.
License
CC0: Public Domain
Who Can Use It
- Machine Learning Engineers: For training and evaluating binary classification models focused on stress detection.
- Signal Processing Researchers: To study the effectiveness of EMD and statistical feature extraction techniques on PPG signals.
- Data Scientists: For statistical analysis and exploration of feature significance in predicting physiological states.
Dataset Name Suggestions
- EMD PPG Stress Feature Set
- Bio-Signal Stress Classification Features
- PPG-EMD Mental Workload Data
- Statistical PPG Stress Indicators
Attributes
Original Data Source: EMD PPG Stress Feature Set
QUALITYLoading...