The output variables are essentially taken from the original dataset. However, a couple of changes were made to make them clean:
- No brackets, dashes or empty spaces
- Only small captions
I did not write out the abreviations, such as time for "t". I found the variables however fairly descriptive. Also there is a tradeoff between length of the measurement name and understanding.
The final data set contains four columns:
- activity - the performed activity when the measurement was taken
- subject - the person who performed the activity coded by an integer
- measurement - the type of measurement that was taken
- mean - the mean of the measurement taken