Skip to content

CodeBook.md

Latest commit

 

History

History
65 lines (53 loc) · 3.4 KB

CodeBook.md

File metadata and controls

65 lines (53 loc) · 3.4 KB

Code Book

This is a document to describe the script inside run_analysis.R

run_analysis.R script is structured as explained below (by comments):

  • Load Libraries
  • Load "Features" and "Activity_labels" Data Sets
  • Load and Manipulate "Test" Data Sets
  • Load and Manipulate "Train" Data Sets
  • Merge "Test" and "Train" Data Sets and Create Tidy Data File
  • Write Tidy Data to TXT

Load Libraries

Load data.table library for reading the data and reshape2 library to use melt() and dcast() operations.

Load "Features" and "Activity_labels" Data Sets

Read second column of "features.txt", which contains the information of the column names of "x_test" and "x_train".

  • For instance, tBodyAcc-mean()-X, tBodyAcc-mean()-Y, tBodyAcc-mean()-Z, etc

Read second column of "activity_labels", which contains the descriptive information of activities

  • WALKING, WALKING_UPSTAIRS, WALKING_DOWNSTAIRS, SITTING, STANDING

Load and Manipulate "Test" Data Sets

  • Read dataset files from UCI HAR Dataset to given name ("test") and prefix ("X", "y" and "subject"). Examples:
    • UCI HAR Dataset/test/X_test.txt
    • UCI HAR Dataset/test/y_test.txt
    • UCI HAR Dataset/test/subject_test.txt
  • Get only columns with mean() or std() in their names using grepl() to subset "x_test"
  • Rename column names of "y_test" and "subject_test"
  • Bind all the 3 data sets by column in the following order "subject_test", "y_test" and "x_test"

Load and Manipulate "Train" Data Sets

  • Read dataset files from UCI HAR Dataset to given name ("train") and prefix ("X", "y" and "subject"). Examples:
    • UCI HAR Dataset/train/X_train.txt
    • UCI HAR Dataset/train/y_train.txt
    • UCI HAR Dataset/train/subject_train.txt
  • Subset "x_train" by columns with mean() or std() in their names
  • Rename column names of "y_train" and "subject_train"
  • Bind all the 3 data sets by column in the following order "subject_train", "y_train" and "x_train"

Merge "Test" and "Train" Data Sets and Create Tidy Data File

  • Define ID variables ("subject", "activity_id", "activity_Label") and measured variables (= remove id_variables) for the melt function
  • Use melt function to convert wide data to long data. Used to calculate mean using dcast() function by:
    • subject
    • activity label ("LAYING","SITTING",etc)
    • features ("tBodyAcc-mean()-X", "tBodyAcc-mean()-Y",etc)

The final data tidy looks like below:

> head(my_tidy_data[, 1:6], n=8)
  subject     activity_Label tBodyAcc-mean()-X tBodyAcc-mean()-Y tBodyAcc-mean()-Z tBodyAcc-std()-X
1       1             LAYING         0.2215982      -0.040513953        -0.1132036      -0.92805647
2       1            SITTING         0.2612376      -0.001308288        -0.1045442      -0.97722901
3       1           STANDING         0.2789176      -0.016137590        -0.1106018      -0.99575990
4       1            WALKING         0.2773308      -0.017383819        -0.1111481      -0.28374026
5       1 WALKING_DOWNSTAIRS         0.2891883      -0.009918505        -0.1075662       0.03003534
6       1   WALKING_UPSTAIRS         0.2554617      -0.023953149        -0.0973020      -0.35470803
7       2             LAYING         0.2813734      -0.018158740        -0.1072456      -0.97405946
8       2            SITTING         0.2770874      -0.015687994        -0.1092183      -0.98682228

Write Tidy Data to TXT

  • Save the "my_tidy_data" data set to "tidy_data.txt" file