The run_analysis.R script contains the script that performs steps 1-5 of the course project.

The script itself is structured so that it includes a section for each step, with comments within the code explaining its purpose and actions.

In the run_analysis.R script I assume the script is located in the parent directory of UCI HAR Dataset directory (the directory that contains the test and train sub-directories as well as related files.

The expected directory structure is as follows:

\run_analysis.R
\UCI HAR Dataset\*.*
\UCI HAR Dataset\test\*.*
\UCI HAR Dataset\train\*.*

• dplyr

1. Load the train and test data into data tables

• 3 files are loaded for each:
  • The data itself (train / test)
  • The Labels (i.e. the Activity codes, numeric range 1:6)
  • The Subjects (i.e. test subject identifiers, as numeric with range 1:30)
• Bind the Labels and Subject columns as new columns of the train and test sets (accordingly)
• Bind the rows of test and train data sets into a single data set, appropriately named allData
• Loaded the descriptive header names and assign them to the allData data set
  • (this was required in step 4, but I decided to do this in this step since it feels more aesthetic to have proper column names on a data set)

I used the apply function to calculate the mean and sd of all measurements columns:

> apply(allData[1:561], 2, mean)  
> apply(allData[1:561], 2, sd)

I restricted the column range to 1:561 since I added (in step 1) two columns (562:563 for Activity and Subject info).

The result of this step is that the mean and standard deviation of every measurement column is printed to the console.

In this step I add the proper values for activities (i.e. mapping the activity codes 1:6 to an activity name like "WALKING", "SITTING" etc.)

This is done using the dplyr library's inner_join function.

The inner join is performed based on identical column names ("Activity") and a identical row counts in the two joined sets.

This was done in step 1, as described above.

Using the dplyr library's aggregate function, I calculate the average (mean) of every measurement, grouped by Activity and Subject.

This means that for 6 activities, and 30 subjects, there will be 180 rows (6x30) in the resulting data set, each containing the average measurements for a specific activity performed by a specific subject.

The resulting data set was saved using write.table function as a tab separated text file named "meanByActivityAndSubject.txt", and uploaded to using the Coursera web site's user interface.