Introduction

Using devices such as Jawbone Up, Nike FuelBand, and Fitbit it is now possible to collect a large amount of data about personal activity relatively inexpensively. These type of devices are part of the quantified self movement – a group of enthusiasts who take measurements about themselves regularly to improve their health, to find patterns in their behavior, or because they are tech geeks. One thing that people regularly do is quantify how much of a particular activity they do, but they rarely quantify how well they do it.

In this project, data from accelerometers on the belt, forearm, arm, and dumbell of 6 participants are being used to predict the manner in which they did the exercise.

Data Preprocessing

library(caret)
## Loading required package: lattice
## Loading required package: ggplot2
library(rpart)
library(rpart.plot)
library(randomForest)
## randomForest 4.6-12
## Type rfNews() to see new features/changes/bug fixes.
library(corrplot)

Download Data

trainUrl <-"https://d396qusza40orc.cloudfront.net/predmachlearn/pml-training.csv"
testUrl <- "https://d396qusza40orc.cloudfront.net/predmachlearn/pml-testing.csv"
trainFile <- "./data/pml-training.csv"
testFile  <- "./data/pml-testing.csv"
if (!file.exists("./data")) {
  dir.create("./data")
}
if (!file.exists(trainFile)) {
  download.file(trainUrl, destfile=trainFile, method="curl")
}
## Warning: running command 'curl "https://d396qusza40orc.cloudfront.net/
## predmachlearn/pml-training.csv" -o "./data/pml-training.csv"' had status
## 127
## Warning in download.file(trainUrl, destfile = trainFile, method = "curl"):
## download had nonzero exit status
if (!file.exists(testFile)) {
  download.file(testUrl, destfile=testFile, method="curl")
}
## Warning: running command 'curl "https://d396qusza40orc.cloudfront.net/
## predmachlearn/pml-testing.csv" -o "./data/pml-testing.csv"' had status 127
## Warning in download.file(testUrl, destfile = testFile, method = "curl"):
## download had nonzero exit status

Read Data

Read the two csv files into two data frames.

trainRaw <- read.csv("C:/Users/Idham/Documents/pml-training.csv")
testRaw <- read.csv("C:/Users/Idham/Documents/pml-testing.csv")
dim(trainRaw)
## [1] 19622   160
dim(testRaw)
## [1]  20 160

Training data set contains 19622 observations and 160 variables, while the testing data set contains 20 observations and 160 variables. The “classe” variable in the training set is the outcome to predict.

Data Cleaning

Clean the data and remove observations with missing values as well as some meaningless variables.

sum(complete.cases(trainRaw))
## [1] 406

First, remove columns that contain NA missing values.

trainRaw <- trainRaw[, colSums(is.na(trainRaw)) == 0] 
testRaw <- testRaw[, colSums(is.na(testRaw)) == 0]

Next, remove of some columns that do not contribute much to the accelerometer measurements.

classe <- trainRaw$classe
trainRemove <- grepl("^X|timestamp|window", names(trainRaw))
trainRaw <- trainRaw[, !trainRemove]
trainCleaned <- trainRaw[, sapply(trainRaw, is.numeric)]
trainCleaned$classe <- classe
testRemove <- grepl("^X|timestamp|window", names(testRaw))
testRaw <- testRaw[, !testRemove]
testCleaned <- testRaw[, sapply(testRaw, is.numeric)]

Cleaned training data set contains 19622 observations and 53 variables, while the testing data set contains 20 observations and 53 variables. The “classe” variable is still in the cleaned training set.

Split Data

Split the cleaned training set into a pure training data set (70%) and a validation data set (30%). We will use the validation data set to conduct cross validation in future steps.

set.seed(22519) # For reproducibile purpose
inTrain <- createDataPartition(trainCleaned$classe, p=0.70, list=F)
trainData <- trainCleaned[inTrain, ]
testData <- trainCleaned[-inTrain, ]

Data Modeling

Fit a predictive model for activity recognition using Random Forest algorithm because it automatically selects important variables and is robust to correlated covariates & outliers in general. We will use 5-fold cross validation when applying the algorithm.

controlRf <- trainControl(method="cv", 5)
modelRf <- train(classe ~ ., data=trainData, method="rf", trControl=controlRf, ntree=250)
modelRf
## Random Forest 
## 
## 13737 samples
##    52 predictor
##     5 classes: 'A', 'B', 'C', 'D', 'E' 
## 
## No pre-processing
## Resampling: Cross-Validated (5 fold) 
## Summary of sample sizes: 10989, 10989, 10991, 10990, 10989 
## Resampling results across tuning parameters:
## 
##   mtry  Accuracy   Kappa      Accuracy SD  Kappa SD   
##    2    0.9908278  0.9883961  0.001675093  0.002121380
##   27    0.9911190  0.9887646  0.001755971  0.002222771
##   52    0.9840572  0.9798290  0.003497420  0.004425063
## 
## Accuracy was used to select the optimal model using  the largest value.
## The final value used for the model was mtry = 27.

Then, estimate the performance of the model on the validation data set.

predictRf <- predict(modelRf, testData)
confusionMatrix(testData$classe, predictRf)
## Confusion Matrix and Statistics
## 
##           Reference
## Prediction    A    B    C    D    E
##          A 1672    1    0    0    1
##          B    5 1129    5    0    0
##          C    0    1 1020    5    0
##          D    0    0   13  949    2
##          E    0    0    1    6 1075
## 
## Overall Statistics
##                                           
##                Accuracy : 0.9932          
##                  95% CI : (0.9908, 0.9951)
##     No Information Rate : 0.285           
##     P-Value [Acc > NIR] : < 2.2e-16       
##                                           
##                   Kappa : 0.9914          
##  Mcnemar's Test P-Value : NA              
## 
## Statistics by Class:
## 
##                      Class: A Class: B Class: C Class: D Class: E
## Sensitivity            0.9970   0.9982   0.9817   0.9885   0.9972
## Specificity            0.9995   0.9979   0.9988   0.9970   0.9985
## Pos Pred Value         0.9988   0.9912   0.9942   0.9844   0.9935
## Neg Pred Value         0.9988   0.9996   0.9961   0.9978   0.9994
## Prevalence             0.2850   0.1922   0.1766   0.1631   0.1832
## Detection Rate         0.2841   0.1918   0.1733   0.1613   0.1827
## Detection Prevalence   0.2845   0.1935   0.1743   0.1638   0.1839
## Balanced Accuracy      0.9983   0.9981   0.9902   0.9927   0.9979
accuracy <- postResample(predictRf, testData$classe)
accuracy
##  Accuracy     Kappa 
## 0.9932031 0.9914024
oose <- 1 - as.numeric(confusionMatrix(testData$classe, predictRf)$overall[1])
oose
## [1] 0.006796941

The estimated accuracy of the model is 99.42% and the estimated out-of-sample error is 0.58%.

Predicting Test Data Set

Apply the model to the original testing data set downloaded from the data source. We remove the problem_id column first.

result <- predict(modelRf, testCleaned[, -length(names(testCleaned))])
result
##  [1] B A B A A E D B A A B C B A E E A B B B
## Levels: A B C D E

Appendix: Figures

  1. Correlation Matrix Visualization
corrPlot <- cor(trainData[, -length(names(trainData))])
corrplot(corrPlot, method="color")

2. Decision Tree Visualization

treeModel <- rpart(classe ~ ., data=trainData, method="class")
prp(treeModel) # fast plot