This repository contains the project course work for Coursera's Getting and Cleaning Data course which is a part of Data Science specialisation.
Given a dataset of human activity as recognised by a Smartphone, we need to process the data to do the following:
- Merge different raw files to form a single data file.
- Make the data file more descriptive with better data labels and representation.
- Create a summary tidy set that summarises all the information on a per subject and activity basis.
The data used for the project has been collected from the accelerometers from the Samsung Galaxy S smartphone. A full description of the raw data is at:
http://archive.ics.uci.edu/ml/datasets/Human+Activity+Recognition+Using+Smartphones
The raw data used for the project can be found at: https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip
- Clone the repository to your R working directory.
- Download the zip file linked above and extract it inside the directory of
this repository. The code assumes that this directory will be called
UCI HAR Dataset. - Execute
run_analysis.R.
You can change the localPath variable at the top of the file to match your
data folder's name. localPath can be set to ./ if the data is in the same
directory as run_analysis.R.
The repository contains the following files:
| File Name | Description |
|---|---|
README.md |
Describes the overall problem and solution structure |
run_analysis.R |
Script that performs analysis and creates the output files. |
CodeBook.md |
Describes the analysis and the variables used to cleanup data. |
tidy.txt |
The final output file after cleaning. |
processedData.txt |
The intermediatary file before summarising data. |
.gitignore |
Standard git file to ignore files not needed |