Gordon CHAN
16 Jul 2015
This is a project to manipulate and tidy up part of the Human Activity Recognition Using Smartphones Data Set that is related to the mean and SD value of the variables.
##Study design and data processing
###Collection of the raw data The original data are from the recordings of 30 subjects performing activities of daily living while carrying a waist-mounted smartphone with embedded inertial sensors.
###Notes on the original (raw) data
The experiments have been carried out with a group of 30 volunteers within an age bracket of 19-48 years. Each person performed six activities (WALKING, WALKING_UPSTAIRS, WALKING_DOWNSTAIRS, SITTING, STANDING, LAYING) wearing a smartphone (Samsung Galaxy S II) on the waist. Using its embedded accelerometer and gyroscope, we captured 3-axial linear acceleration and 3-axial angular velocity at a constant rate of 50Hz. The experiments have been video-recorded to label the data manually. The obtained dataset has been randomly partitioned into two sets, where 70% of the volunteers was selected for generating the training data and 30% the test data.
The sensor signals (accelerometer and gyroscope) were pre-processed by applying noise filters and then sampled in fixed-width sliding windows of 2.56 sec and 50% overlap (128 readings/window). The sensor acceleration signal, which has gravitational and body motion components, was separated using a Butterworth low-pass filter into body acceleration and gravity. The gravitational force is assumed to have only low frequency components, therefore a filter with 0.3 Hz cutoff frequency was used. From each window, a vector of features was obtained by calculating variables from the time and frequency domain.
##Creating the tidy datafile
###Guide to create the tidy data file The tidy file is created through the following steps:
- Download the original dataset
- Load the following componenst of the dataset:
- Lables for the features of the variables
- Lables for the type of activities
- For the Test and Train data sets:
- Subject ID
- Activity index
- Main values dataset
- Merge the three components of the Test and Train data sets
- Merge the Test and Train data sets into a single data set
- Select the Subject ID, Activity index, and columns related to mean and SD of the variables into a data frame
- Clean the data as described by the next section
###Cleaning of the data The data is cleaned throguh the following steps:
- The data frame is sorted by the subject ID and then by the activity index
- The activity index is replaced by the the actual activities
- A wide format data frame is created by aggregating the mean for all the trials by each subject on each activity
- The wide data frame is melted to form a narrow data frame
- The variables are renamed according to the elements in the original name
- Each variable is classified as the mean or SD of a measurement
- The data frame is recasted to show the mean and SD of each measurement per entry(row)
The resulting data frame is tidy and clean:
a) Each column corresponds to a variable (each feature) b) Each observation (mean and SD of each feature) is in a different row c) There is a single data table
Sample from the data set:
head(data)##
## 1 function (..., list = character(), package = NULL, lib.loc = NULL,
## 2 verbose = getOption("verbose"), envir = .GlobalEnv)
## 3 {
## 4 fileExt <- function(x) {
## 5 db <- grepl("\\\\.[^.]+\\\\.(gz|bz2|xz)$", x)
## 6 ans <- sub(".*\\\\.", "", x)
tail(data)##
## 160 if (!found)
## 161 warning(gettextf("data set %s not found", sQuote(name)),
## 162 domain = NA)
## 163 }
## 164 invisible(names)
## 165 }
##Description of the variables in the narrow_data.txt file General description of the file:
data <- read.table("narrow_data.txt", header=TRUE, stringsAsFactors=FALSE)- Dimensions of the dataset
dim(data)## [1] 5940 5
The dataset have 5 columns and 5940 rows.
- Summary of the data
summary(data)## subject activity feature mean
## Min. : 1.0 Length:5940 Length:5940 Min. :-0.99762
## 1st Qu.: 8.0 Class :character Class :character 1st Qu.:-0.93140
## Median :15.5 Mode :character Mode :character Median :-0.12974
## Mean :15.5 Mean :-0.30898
## 3rd Qu.:23.0 3rd Qu.:-0.01192
## Max. :30.0 Max. : 0.97451
## std
## Min. :-0.9977
## 1st Qu.:-0.9714
## Median :-0.9194
## Mean :-0.6597
## 3rd Qu.:-0.3638
## Max. : 0.6871
- Variables present in the dataset
names(data)## [1] "subject" "activity" "feature" "mean" "std"
###Variable 1: subject Subject ID is unique identifier to the subjects.
- Class of the variable: integer
- Unique values/levels of the variable: 1 to 30
- Unit of measurement: no unit
####Notes on variable 1: The subjects are a group of 30 volunteers between 19-48 years of age.
###Variable 2 : activity Activity is description to the activity performed by the subeject while wearing the smartphone at their waist.
- Class of the variable: character
- Unique values/levels of the variable:
activity <- data.frame(unique(data$activity))
colnames(activity) <- "List of Activities"
print(activity)## List of Activities
## 1 LAYING
## 2 SITTING
## 3 STANDING
## 4 WALKING
## 5 WALKING_DOWNSTAIRS
## 6 WALKING_UPSTAIRS
- Unit of measurement: no unit
###Variable 3 : feature feature is description of the processed signals from the accelerometer and the gyroscope of the smartphone.
- Class of the variable: character
- Unique values/levels of the variable:
feature <- data.frame(unique(data$feature))
colnames(feature) <- "List of Features"
print(feature)## List of Features
## 1 function_domain_body_acceleration_jerk_signal_magnitude
## 2 function_domain_body_acceleration_jerk_x-axis_signal
## 3 function_domain_body_acceleration_jerk_y-axis_signal
## 4 function_domain_body_acceleration_jerk_z-axis_signal
## 5 function_domain_body_acceleration_signal_magnitude
## 6 function_domain_body_acceleration_x-axis_signal
## 7 function_domain_body_acceleration_y-axis_signal
## 8 function_domain_body_acceleration_z-axis_signal
## 9 function_domain_body_gyroscopic_jerk_signal_magnitude
## 10 function_domain_body_gyroscopic_signal_magnitude
## 11 function_domain_body_gyroscopic_x-axis_signal
## 12 function_domain_body_gyroscopic_y-axis_signal
## 13 function_domain_body_gyroscopic_z-axis_signal
## 14 time_domain_body_acceleration_jerk_signal_magnitude
## 15 time_domain_body_acceleration_jerk_x-axis_signal
## 16 time_domain_body_acceleration_jerk_y-axis_signal
## 17 time_domain_body_acceleration_jerk_z-axis_signal
## 18 time_domain_body_acceleration_signal_magnitude
## 19 time_domain_body_acceleration_x-axis_signal
## 20 time_domain_body_acceleration_y-axis_signal
## 21 time_domain_body_acceleration_z-axis_signal
## 22 time_domain_body_gyroscopic_jerk_signal_magnitude
## 23 time_domain_body_gyroscopic_jerk_x-axis_signal
## 24 time_domain_body_gyroscopic_jerk_y-axis_signal
## 25 time_domain_body_gyroscopic_jerk_z-axis_signal
## 26 time_domain_body_gyroscopic_signal_magnitude
## 27 time_domain_body_gyroscopic_x-axis_signal
## 28 time_domain_body_gyroscopic_y-axis_signal
## 29 time_domain_body_gyroscopic_z-axis_signal
## 30 time_domain_gravity_acceleration_signal_magnitude
## 31 time_domain_gravity_acceleration_x-axis_signal
## 32 time_domain_gravity_acceleration_y-axis_signal
## 33 time_domain_gravity_acceleration_z-axis_signal
- Unit of measurement: no unit
- Schema used to constructed the entries:
The names are construted through a custom function which look for naming elements from the orignal variable name:
| Sequence | Element from orinal name | Elements used in new name |
|---|---|---|
| 1 | t or f | time domain or function domain |
| 2 | body or gravity | body or gravity |
| 3 | Acc or Gyro | Acceleration or Gyroscopic |
| 4 | Jerk or nil | Jerk or nil |
| 5 | X or Y or Z | x-axis or y-axis or z-axis |
| 6 | for all variables | signal |
| 7 | Mag | magnitude |
###Variable 4 : mean mean is the mean of the mean measurement obtained from all of the trials.
- Class of the variable: numeric
- Unit of measurement: no unit
###Variable 5 : std std is the mean of the standard deviation measurement obtained from all of the trials
- Class of the variable: numeric
- Unit of measurement: no unit
##Sources Davide Anguita, Alessandro Ghio, Luca Oneto, Xavier Parra and Jorge L. Reyes-Ortiz. Human Activity Recognition on Smartphones using a Multiclass Hardware-Friendly Support Vector Machine. International Workshop of Ambient Assisted Living (IWAAL 2012). Vitoria-Gasteiz, Spain. Dec 2012
Description of the dataset can be found from: https://archive.ics.uci.edu/ml/datasets/Human+Activity+Recognition+Using+Smartphones
Dataset can be downloaded from: https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip