Skip to content

CodeBook.md

Latest commit

 

History

History
42 lines (24 loc) · 1.79 KB

CodeBook.md

File metadata and controls

42 lines (24 loc) · 1.79 KB

Code Book - Getting and Cleaning Data Project

Original Data

The original data used in this project is available here.

A full description of all files and variables is included with the data set. The data files that particularly concern this project are as follows.

  • 'features.txt': List of all features.

  • 'activity_labels.txt': Links the class labels with their activity name.

  • 'train/X_train.txt': Training set.

  • 'train/y_train.txt': Training labels.

  • 'test/X_test.txt': Test set.

  • 'test/y_test.txt': Test labels.

  • 'train/subject_train.txt': Each row identifies the subject who performed the activity.

  • 'test/subject_test.txt': Each row identifies the subject who performed the activity.

Data Tidying

Using run_analysis(), the above sets were combined to form one data set, which relates activities, values and subjects. This data set is then filtered to include on those values which are a mean or standard deviation. Averages are then taken after grouping by subject and activity. Finally, the columns are renamed to a meaningful, human-readable list of headings. This results in a data set with 180 rows (30 subjects and 6 acitivities), and 82 columns. The columns are as follows.

  • Column 1 : Activity name
  • Column 2 : Subject
  • Column 3:82 : Average values of measurements for each activity and each subject.

The following rules were applied to tidy the column names.

  • t at start of line, replaced with Time
  • f at start of line, replaced with Frequency
  • (), -, replaced with empty string
  • mean, replaced with Mean
  • std, replaced with Std

For example, the variable tBodyAcc-mean()-X from the original data set is replaced with TimeBodyAccMeanX in the final tudy data set.