Codebook for Getting And Cleaning Data Course Project
How the code works: This code takes data from this web site: http://archive.ics.uci.edu/ml/datasets/Human+Activity+Recognition+Using+Smartphones It then processes it into a tidy data set as defined by John Hopkins Getting and Cleaning Data class: https://class.coursera.org/getdata-008 The full data set usedfor writing this particular assignment was downloaded on 9/3/2014 from ... https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip and may have changed by the time you read this. (I had taken a previous class but did not get credit for the initial upload to githib so any references you see to class "getdata-007" should be changed to "getdata-008").
Machine information: Windows 8.1
R Version 3.1.10 "Sock it to Me"
64 bit
The code reads the text file data for the test and training sets. It merges the two sets and sets the expected column names. A tidy data file is created with standard deviation and mean for each activity and participant in the study.
Variable names: The training and test sets for x and y are stored in "trainx", "testy", etc... The subjects are stored in "testsub" and "trainsub".
The 561 features used this in data set are stored in "features".
The assignment calls for the test and train data sets to be merged. This concatenation is done with "testbounddata" and "trainbounddata" and the combined data is stored in the "data" variable.
The GREP command to extract the columns needed for mean and standard deviation is a bit complicated. The command is this:
dataset1<-data[,c(1,2 ,grep(".-std.",colnames(data)), grep(".-Std.", colnames(data)),grep(".-Mean.", colnames(data)), grep(".-mean.", colnames(data)))]
There is a good explanation of this here: http://www.regular-expressions.info/rlanguage.html but the point is to select the columns with variants of "std" for standard deviation and "mean" in them for creating the final data set.
The reshape2 library is used for melting and dcasting the data. Hadley Wickham (2007). Reshaping Data with the reshape Package. Journal of Statistical Software, 21(12), 1-20. URL http://www.jstatsoft.org/v21/i12/.
The code takes the full data set and sets the column names for the six exercises from the study. There is some optional code to save a data set mentioned in the exercise but is not actually needed to be saved. You can uncomment that line if you want the extra dataset.
Finally, all the columns except the name of the specific exercise are extraces and melted into a data frame with the reshape2 melt command. The means (averages) for each column are computed and written to "FinalDataSet.csv" in CSV format.