README

Getting and Cleaning Data - Course Project

This text explains how all of the scripts work and how they are connected.

Data

The data for this course project are available from Dataset.zip

You can read each of the text files using the read.table() function in R. For example, reading in 'features.txt' (from the UCI HAR Dataset folder) can be done with the following code:

features <- read.table("./UCI HAR Dataset/features.txt")

R packages required

The packages required to run the script are as follows:

• plyr
• reshape2
• tidyr

Procedure:

Set the working directory to the folder in which the UCI HAR Dataset folder is saved.

Step I - Merge the training and the test sets to create one data set

1. Read the list of all features

    features <- read.table("./UCI HAR Dataset/features.txt", 
                           col.names = c("Feature_Num", "Feature_Name"), 
                           colClasses = "character")
   
    features[,2] <- gsub("\\()", "", features[,2])
   
    features[,2] <- gsub("-|,|\\(|\\)", "_", features[,2])
   

2. Read the training data set

    X_train <- read.table("UCI HAR Dataset/train/X_train.txt", 
                          colClasses = "numeric", 
                          col.names = features[,2])
   

3. Read the training labels

    y_train <- read.table("UCI HAR Dataset/train/y_train.txt", 
                          col.names = "Activity_Label")
   

4. Read the training subjects' identifiers

    subject_train <- read.table("UCI HAR Dataset/train/subject_train.txt", 
                                col.names = "Subject")
   

5. Create a data frame for training data

    training <- cbind(subject_train, y_train, X_train)
   

6. Read the test set

    X_test <- read.table("UCI HAR Dataset/test/X_test.txt", 
                         colClasses = "numeric", 
                         col.names = features[,2])
   

7. Read the test labels

    y_test <- read.table("UCI HAR Dataset/test/y_test.txt", 
                         col.names = "Activity_Label")
   

8. Read the test subjects' identifiers

    subject_test <- read.table("UCI HAR Dataset/test/subject_test.txt", 
                               col.names = "Subject")
   

9. Create a data frame for test data

    test <- cbind(subject_test, y_test, X_test)
   

10. Combine the two data frames: training and test

    comb_data <- rbind(training, test)
    

Step II - Extract only the measurements on the mean and standard deviation for each measurement

1. Get the index of the measurement on the mean and standard deviation

    col_index <- c(1, 2, grep("*[Mm]ean|*[Ss]td*", names(comb_data)))
    col_index <- col_index[! col_index %in% c(557:563)]
   

2. Extract the data

    extr_data <- comb_data[, col_index]
   

Step III - Uses descriptive activity names to name the activities in the data set

1. Read the activity labels with their activity name

    activity_labels <- read.table("./UCI HAR Dataset/activity_labels.txt", 
                                  col.names = c("Activity_Label", "Activity"))
   

2. Update extr_data with descriptive activity names

    extr_data_upd <- join(activity_labels, extr_data, by = "Activity_Label")
    
    extr_data_upd$Activity_Label <- NULL
   

3. Move Subject column to the first of the data frame extr_data_upd

    extr_data_upd <- cbind(extr_data_upd$Subject, extr_data_upd)
    extr_data_upd$Subject <- NULL
    names(extr_data_upd)[1] <- "Subject"
   

Step IV - Appropriately label the data set with descriptive variable names.

This has been done in Step I. The descriptive values listed in features.txt would be appropriate. They are slightly modified and used as the variable names for the data set.

Step V - From the data set in Step IV, create a second, independent tidy data set with the average of each variable for each activity and each subject

1. Split the data set by Subject and by Activity

    sp_list <- split(extr_data_upd, list(extr_data_upd$Subject,extr_data_upd$Activity))
   

2. Calculate the average of each variable for each subject and each activity

    avg_mat <- t(sapply(sp_list, function(df) {colMeans(df[, 3:81])}))
   

3. Transform the matrix avg_mat into a data frame avg_df

    avg_df <- as.data.frame(avg_mat)
   

4. Update the column names

    colnames(avg_df) <- paste("mean(", colnames(avg_df), ")", sep = "")
   

5. Extract the row names of avg_df

    subject_activity <- rownames(avg_df)
   

6. Update the data frame avg_df by adding the row names as a column

    avg_df_upd <- cbind(subject_activity, avg_df)
   

7. Get the tidy data set

    tidy_data <- separate(avg_df_upd, subject_activity, 
             			  into = c("Subject", "Activity"), 
    					  sep = "\\.")
   

8. Get the final tidy data set

    final_tidy_data <- tidy_data[order(as.numeric(tidy_data$Subject)), ]
    rownames(final_tidy_data) <- NULL
   

Step VI - Export the data set into a text file

write.table(final_tidy_data, 
            file = "final_tidy_data.txt", 
            sep = "\t", 
            row.names = FALSE)