Getting and Cleaning Data: Course Project

Summary

The objective of this analysis was to collect, work with, and clean a data set, in order to prepare a tidy data set that can be used for later analysis. Data is collected from a group of 30 subjects (age from 19-48 years) using Samsung Galaxy S smartphone.

The raw data is available here: https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip

The original source of the data is available below: http://archive.ics.uci.edu/ml/datasets/Human+Activity+Recognition+Using+Smartphones

The 30 subjects have to perform 6 different physical activities. These activites include walking, walking upstairs and downstairs, sitting, standing and laying. The smartphones track volunteers' data from their waist using the phones' accelerometer and gyroscope.

In short, the data would be extracted, processed (clean and tidy) and transformed into a new tidy data set.

run_analysis.R

The file run_analysis.R in this repository is used to perform the following:

• Initial setup

  • Check if the raw data set has been downloaded / unzipped in the current directory.
  • If not, the script will download raw data and unzip it

• Merges the training and the test sets to create one data set

• Extracts only the measurements on the mean and standard deviation for each measurement

• Name the activities in the data set with descriptive names

• Labels the data set with descriptive variable names

• Get the average of each variable for each activity and subject

• Wrtie the clean data as tidyData.txt in the current directory.

Running run_analysis.R

To run the run_analyis.R script, you do not need to setup anything. There is no prerequisite. Just download the script and run it. To run the script, load it into your RStudio and hit the Source to execute. Alternatively, you can select all the script content in RStudio and hit the keyboard button Ctrl+Enter.

The script will check if the raw dataset is available in the current working directory. Else, the script will download and unzip it. The working files and the tidy data output [tidyData.txt] will be available in your current working directory getwd().

You can also see the location of the tidyData.txt printed at the end of the console after the script finishes running.

All other procedures are commented in the script.

Approach

Once the raw dataset is available in your current directory, the training and test data sets are loaded and then merged into one single data set using merge. The data sets contain alot of information but we are only interested in certain data. Cleaning is done on this data set to ensure that we only extract observations with the mean and standard deviation.

The data set is also labelled with a more descriptive column names using grep and gsub functions in R. Lastly, aggregate function is applied on the data set to find the average of each variable for each activity and subject.

The final clean data is saved as tidydata.txt using write.table R function. Also all the variables will be removed after the script has completed running.

CodeBook

The CodeBook.md file, which can be found in this repository, explains the transformations performed and the resulting data and variables.