Analysis run_analysis.R solves a course project for getting and cleaning data - https://www.coursera.org/course/getdata. This readme describes how run_analysis.R works.
The script is divided into 5 sections, just mirroring what is taked in the course project.
On this step we merge training and test data into one. For merging we use following source. There are 2 folders: test and train. From test folder we take 3 files: x_test.txt, subject_test.txt, y_test.txt. Same three files are taken from train folder respectively.
As result we bind rows into 3 new data frames and keep for later use.
There are quite many features measured, we are interested only in those that contain mean or std in their names. To solve this step we use features.txt that has names of columns (measurements) that are used in the x_test.txt and x_train.txt.
As result we remove columns that we are not interested in from the x_* data.frame.
Activities are recorded as identifiers in the y_* files. However there is activities.txt that has a match beween identifier and a descriptive activity name (SITTING, WALKING etc).
As result of this we have meaningful names of activities in our in-memory data.frames.
At this step we bind together columns: activities, measurements, and activities.
At this step we are aggregating column means by subject and activity.