CodeBook.md

CodeBook

Attribute information

The source of the original data describes the data as follows.

For each record in the dataset it is provided:

• Triaxial acceleration from the accelerometer (total acceleration) and the estimated body acceleration.
• Triaxial Angular velocity from the gyroscope.
• A 561-feature vector with time and frequency domain variables.
• Its activity label.
• An identifier of the subject who carried out the experiment.

More information can be found at the original data publisher's website here.

Transformations on the original data:

The original data comes in separate files for subject, activity and the actual measurements (variables):

Subject data files:

• test/subject_test.txt and
• train/subject_train.txt

Activity data files:

• test/y_test.txt and
• train/y_train.txt

Additionally, the labels for the activities are available in the data file activity_labels.txt in the root directory.

The variables are available in the following data files:

• test/X_test.txt and
• train/X_train.txt

Again with descriptors available in another file called features.txt.

The original data is partitioned by test and train data sets for use with machine learning algorithms, but for our use we just merge both partitions of the same data every time, as requested by the instructions for this assignment.

We then choose the columns that correspond to mean and standard deviation measurements and filter out the remaining measurement columns.

The last steps are to save the tidy data set we just created and create and save another one containing the same data, but grouped by subject and activity.