The purpose of this R project is to provide a tidy data set for later analysis from the source data of this smartphone trial set: http://archive.ics.uci.edu/ml/datasets/Human+Activity+Recognition+Using+Smartphones
##Setup Download and unzip the raw data to your working directory. The unzipped directory of raw data should be titled: "UCI HAR Dataset"
Download the run_analysis.R to your working directory
##Running the script The run_analysis.R contains a run_analysis function that when called returns the tidy data set
To run the script just type the following into the console
source("run_analysis.R")
df <- run_analysis()
*Note the example drops the tidy data into a new data.frame df for you to manipulate at will. If you prefer to simply return the dataset to the console, you can just run run_analysis().
I have included the code below with comments for each of the sections
##Code run_analysis <- function() {
##Loads training data and adds the subjects and activity labels to the data frame
trainingData <- read.table("./UCI HAR Dataset/train/X_train.txt")
trainingLabels <- read.table("./UCI HAR Dataset/train/y_train.txt")
trainingData$Activity = c(t(trainingLabels))
trainingSubjects <- read.table("./UCI HAR Dataset/train/subject_train.txt")
trainingData$Subject = c(t(trainingSubjects))
##Loads test data and adds the subjects and activity labels to the data frame
testData <- read.table("./UCI HAR Dataset/test/X_test.txt")
testLabels <- read.table("./UCI HAR Dataset/test/y_test.txt")
testData$Activity = c(t(testLabels))
testSubjects <- read.table("./UCI HAR Dataset/test/subject_test.txt")
testData$Subject = c(t(testSubjects))
##Merges the training and the test sets to create one data set.
mergedData <- rbind(trainingData, testData)
##Loads the variable labels
varLabels <- read.table("./UCI HAR Dataset/features.txt")
##Sets the raw variable names on the merged data set
names(mergedData) <- c(as.character(varLabels[,2]), "Activity", "Subject")
##Extracts only the measurements on the mean and standard deviation for each measurement.
trimmedData <- mergedData[,c(names(mergedData)[grep("ean", names(mergedData))] ,names(mergedData)[grep("std", names(mergedData))], "Activity", "Subject")]
##Loads the activity keys for using human readable activity descriptions
activityKeys <- read.table("./UCI HAR Dataset/activity_labels.txt")
##Creates a list split by subjects and activities
bySubjectByActivity <- split(trimmedData, list(trimmedData$Subject, trimmedData$Activity))
##calculates the mean for each variable by user nad by activity
meansBySByA <- lapply(bySubjectByActivity, function(x) colMeans(x[,names(trimmedData[, c(-87, -88)])]))
##Builds the tidy data frame
for(i in 1:30) {
for(j in 1:6) {
if(i == 1 & j == 1)
{
tidyData <- data.frame(matrix(unlist(meansBySByA[["1.1"]][1:86]), nrow=1, byrow=T))
names(tidyData) = names(meansBySByA[["1.1"]][])
tidyData$subject = 1
##Uses descriptive activity names to name the activities in the data set
tidyData$activity = as.character(activityKeys[activityKeys$V1 == j, 2])
} else {
listName <- paste(i, j, sep=".")
tempDf <- data.frame(matrix(unlist(meansBySByA[[listName]][1:86]), nrow=1, byrow=T))
names(tempDf) = names(meansBySByA[[listName]][])
tempDf$subject = i
##Uses descriptive activity names to name the activities in the data set
tempDf$activity = as.character(activityKeys[activityKeys$V1 == j, 2])
tidyData <- rbind(tidyData, tempDf)
}
}
}
##Appropriately labels the data set with descriptive activity names.
##Removes the ()
names(tidyData) = gsub("\\(\\)", "", names(tidyData))
##Removes the -
names(tidyData) = gsub("-", ".", names(tidyData))
##Returns the tidy data set
tidyData
}