Skip to content

Latest commit

 

History

History
49 lines (39 loc) · 3.44 KB

README.md

File metadata and controls

49 lines (39 loc) · 3.44 KB
Raw

Overview

This is a simple R script for reading, merging, and cleaning up the data set located at https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip In particular, it:

  • Downloads/extracts the data set (assuming they haven't been downloaded/extracted already). This creates a directory named UCI HAR Dataset with the original data
  • Merges together the training and testing sub-datafiles
    • The main directory contains test and train subdirectories, activity_labels.txt, features.txt, features_info.txt, and a README at README.txt.
    • activity_labels.txt has each line with a number and a corresponding activity. During each observation/test, the individual is performing one of these activities.
    • features.txt contains the original names of the 561 calculations that were performed on each individual each time they were tested.
    • features_info.txt contains helpful descriptions pertaining to how the names of the variables used in features.txt were arrived at.
    • Training set is located inside the "train" directory and the testing set is located in the test data set.
    • Inside the directories, there are 3 files: subject_[train/test].txt, X_[train/test].txt, and y_[train/test].txt. The "Inertial Signals" subdirectory in each is not used in this analysis.
    • subject_[train/test].txt has a single number on each line corresponding to the indivual who was being tested. This is a subject number identifier.
    • X_[train/test].txt has each line being the set of numeric calculations performed. There are 561 separate calculations on each line corresponding to the names from features_info.txt
    • y_[train/test].txt has each line corresponding to the activity that was being performed each time. They are represented here by a number.
    • read.table is used to combine the X training and testing sets together, then the columns are changed to the values from features.txt
    • cbind is then used to combine this data together with the subject and activities for each test.
    • Performed inside readData function.
  • Extracts only the measurements on the mean and standard deviation for each measurement.
    • This is performed using grep inside extractData function.
  • Uses descriptive activity names to name the activities in the data set
    • Performed by setGoodNames function using R sub function to replace pertions of the activity names
  • Appropriately labels the data set with descriptive variable names.
    • Variable names chosen to describe them clearly as neccesary
  • From the data set in step 4, creates a second, independent tidy data set with the average of each variable for each activity and each subject.
    • Performed by getMeanData function using R aggregate function to generate the mean of each variable for each activity and subject.

Installation

Installation is not neccesary, per se as this is a script.

If you are working within R/RStudio then you can simply download "run_analysis.R" and run

source("run_analysis.R")

Usage

source("run_analysis.R")

Output

This will generate 3 values in the global environment: myOriginalData (original dataset), extractedData (the dataset with mean and standard deviation calculations extracted), and meanCalculationsData (the mean of each variable from the extracted data for each subject and activity).

Note

In the data set, I have excluded the features that mention meanFreq() as the description of how those were performed does not match a normal simple mean calculation.