Skip to content

Latest commit

 

History

History
23 lines (14 loc) · 1.72 KB

CodeBook.md

File metadata and controls

23 lines (14 loc) · 1.72 KB
Raw

Code Book

The averages.csv file consists of 4 variables as follows:

  • subject_id : the id of the subject who wore the accelerometer
  • activity : the activity being undertaken while the accelerometer was being worn (a factor with 6 levels: WALKING, WALKING_UPSTAIRS, WALKING_DOWNSTAIRS, SITTING, STANDING, LAYING)
  • variable : the variable being measured by the accelerometer
  • mean : the average measurement of the variable for a particular subject and activity

Method

Having downloaded and expanded the UCI HAR Dataset, the features file was read, and the interesting features were recorded as the ids of those features whose name contained either "mean" or "std".

The Training accelerometer data was read as a series of fixed width (16 character) columns, and all columns which were not in the interesting features were discarded. This remaining data was then combined with the corresponding subject and activity data to produce a combined data set.

The above steps were repeated for the Test accelerometer dataset, and the two dataset's were combined to produce a single dataset containing all the accelerometer observations for all subjects and activities.

The activity column was converted to a factor using the activity labels, and the data was then melted by subject and activity to produce a tidy data set with one variable measurement per row. This was then saved to the mean_and_std_obvs.csv file.

The final step was to use the ddply function to split the measurements into groups by a combination of subject, activity and variable being measured, calculate the average of the resulting groups, and then recombine into the final dataset, which was then written to file as averages.csv.