runAnalysis.R

cleanData<-function(){
  
    traindata <- read.table("./train/X_train.txt")
    traindata_labels <- read.table("./train/y_train.txt")
    subject_train <- read.table("./train/subject_train.txt") # Loading the Training data and also subject and activity labels
    
    
    testdata <- read.table("./test/X_test.txt")
    testdata_labels <- read.table("./test/y_test.txt")
    subject_test <- read.table("./test/subject_test.txt") # Loading the Testing data and also subject and activity labels 
    
    
    myData <- rbind(traindata,testdata)    #merging training and test measurements
    myLabels <- rbind(traindata_labels,testdata_labels) # merging training labels and testing labels
    mySubject <- rbind(subject_train,subject_test)  # merging training subject id and testing subject id
    
    variables <- read.table("./features.txt")   #reading the feature names

    variable_names <- variables$V2
    
    colnames(myData) <- variable_names
    
    cut_mean_std <- grep("[Mm][Ee][Aa][Nn]|[Ss][Tt][Dd]"
                         ,colnames(myData))                   # Extracting only the mean and std measurements
     
    myData_meanstd <- myData[,cut_mean_std]
    
    names <- colnames(myData_meanstd)
    
    names = gsub("tBodyAcc","linearaccbody(Time)",names)
    names = gsub("tGravityAcc","gravityaccbody(Time)",names)
    names = gsub("Mag","magnitude",names)
    names = gsub("tBodyGyro","angvelocitybody(Time)",names)           # Renaming variable names for better understanding
    names = gsub("fBodyAcc","linearaccbody(Frequency)",names)
    names = gsub("fBodyGyro","angvelocitybody(Frequency)",names)
    names = gsub("fBodyBodyGyro","BodyBodyangvelocity(Frequency)",names)
    names = gsub("fBodyBodyAcc","BodyBodylinearacc(Frequency)",names)
    
    colnames(myData_meanstd)= names
    
    cleaned_data <- myData_meanstd
        
    cleaned_data$"Activity Label" <- myLabels
    cleaned_data$"Subject No." <- mySubject
    #cleaned_data = cleaned_data[,1:88]
    return (cleaned_data)
}

createDataset<-function(dataset){
  
  col_names = colnames(dataset)
  data_split <- split(dataset,dataset$"Subject No.")     #Split Dataset according to Subject
  
  func <- function(data=data_split){
    m <- split(data,data$"Activity Label")
    l <- sapply(m,colMeans)                  #function for taking mean of each activity
    return (l)}
  
  gdata <- sapply(data_split,func)   #calculating mean for each activity for each subject
  
  for (i in 1:30){
    n <- rbind(gdata[1:88,i],gdata[89:176,i],gdata[177:264,i],gdata[265:352,i],gdata[353:440,i],gdata[441:528,i])  
    #creating variable for storing data for each subject
    assign(paste0("n",i),n)
    
  }
  
  new_dataset <- rbind(n1,n2,n3,n4,n5,n6,n7,n8,n9,n10,n11,n12,n13,n14,n15,n16,n17,n18,n19,n20,n21,n22,n23,n24,n25,n26,n27,n28,n29,n30)
  new_dataset <- data.frame(new_dataset) #merging data of all sujects to create one dataset
  
  colnames(new_dataset) <- col_names
  
  activity_lables <- c("SUBJECT IS WALKING ON LEVELLED SURFACE",
   "SUBJECT IS WALKING UPHILL OR UPSTAIRS",
   "SUBJECT IS WALKING DOWNHILL OR DOWNSTAIRS",       #Descriptive activity labels
   "SUBJECT IS SITTING",
   "SUBJECT IS JUST STANDING",
   "SUBJECT IS LAYING OR RESTING")
  
  new_dataset$"Activity Label Description" <- new_dataset$"Activity Label" 
  
  for (i in 1:6){
    d_labels <- gsub(i,activity_lables[i],new_dataset$"Activity Label Description")
    new_dataset$"Activity Label Description" <- d_labels
    
  }
  
  #new_dataset$"Activity Label" <- descriptive_labels
  
  write.csv(new_dataset,"./tidy_dataset.csv")
  
  write.table(new_dataset,"./tidy_dataset.txt",row.names = FALSE)   #writing data to text file 
  
  
}