Getting and Cleaning Data Course Project

This is the course project for the coursera.org course "Getting and Cleaning Data".

Project contents

• README.md: this file.

• run_analysis.R: R code for the analytics. Usage described in next section.

• CodeBook.md: Description for the run_analysis.R code.

Please note that the project does not include the data on which run_analysis.R works on.

Purpose

This project is intented for the cleaning of the data contained in

     https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip

A description of these data can be found here:

     http://archive.ics.uci.edu/ml/datasets/Human+Activity+Recognition+Using+Smartphones

Internal working

First, the files under the "train" and "test" read, and merged. The data about means and standard deviations is extracted from the merged "X_*.txt" files.

Then, "clean" (lowercase, no symbols except commas and hyphens) labels are produced out of the "features.txt" file, and added to the columns.

In a similar way, activity labels are also cleaned and assigned to the data in the "y_.txt" files.

Finally, the subject information, taken from the "subject_*.txt" files, the activity labels, and the mean and standard deviation data are merged into a data frame.

This data frame is saved into the "tidy_data.csv" file.

In a second stage, all the data specific for each feature for each subject and activiy combination is averaged, and written to the "tidy_data_means.csv" file.

Usage

The raw data to be loaded is relatively small (around 50 mb), so memory should not be a concern unless you are working on a very old machine.

1. Use git to clone this project onto your computer:

git clone https://github.com/ricrogz/getdata_CP

2. Download the dataset on which this project works from:

https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip

3. Unzip the downloaded file so that the "UCI HAR Dataset" directory resides in the same directory created in step 1. This is necessary since the run_analysis.R expects to find the data under this path.

4. Start up R or Rstudio, set the working path to the project's directory, source the run_analysis.R file, and and call the run_analysis() function to load the data, tidy it up, and return a data frame with the clean data.