##README.md

getting-data-project repository - an assignment for the Coursera "Getting and Cleaning Data" course.

###Contents

1. run_analysis.R - an R script to analyze Samsung data (Human Activity Recognition Using Smartphones). This script reads raw data files from the UCI HAR study and produces a tidy data text file called "UCI-HAR-tidy-data.txt". The output file can be manipulated in R or viewed directly using a spreadsheet program such as MS Excel or LibreOffice Calc.

   The steps in the analysis are:

   • Read the raw data files into data tables - X_test.txt, X_train.txt (test and training sets of the measurements), y_test.txt, y_train.txt (activity codes), subject_test.txt, subject_train.txt (subject IDs). Also read activity_label.txt, which gives the list of activity labels, and features.txt, which gives the variable names for the 561 variables in the raw data.
   • Merge the test and training sets (for X, y and subject) by appending rows.
   • Assign text labels for activities and column names, from activity_label.txt and features.txt.
   • Add subject IDs and activity labels to the table of measurements.
   • Select the columns for analysis: here we just want the means and standard deviations of the various measurements.
   • Group measurements by subject and activity.
   • Find the average value of each variable for each combination of subject and activity.
   • Write the resulting table to a text file.

2. CodeBook.md - a description of the variables in the tidy data produced by run_analysis.R.

3. UCIdocs - a folder containing documentation about the experiment and the raw data format, from the UCI Human Activity Recognition study.