Extracted Feature Category | Extracted Feature Name | Description/Rationale |
---|---|---|
Measures of central tendency | Mean | Measures of central tendency describe the center or typical value of a dataset. The mean is the average value of the data which helps us capture the general level of the data and provide information about the baseline behavior. |
Measures of dispersion | Standard deviation (SD) | Measures of dispersion describe how spread out the data are. The SD measures how much the data varies from the average value. |
Minimum (Min) & Maximum (Max) | Min and Max values give the lower and upper range of the data which can be beneficial when certain ranges of values are indicative of specific conditions (e.g., pain severity/intensity) | |
Median deviation (MD) | The median deviation is a measure of how much the data varies from the middle value [26]. We extract median deviation (MD) related features which consist of mean-MD, max-MD, and min-MD so we can capture variability while being less influenced by extreme values (e.g., high light levels at night show high MD feature, but low Mean compared to daytime). | |
Measures of shape | Slope | Slope measures the shape (or pattern) of the data distribution by measuring how steeply a line fits to the data. This feature is important for capturing trends, identifying periods of growth or decline. |
Mean-crossing-rate (MCR) | The MCR measures how often the data goes above and below its average value [27]. This feature is useful for characterizing oscillatory behavior and cyclical patterns in the data. |