Pratical Machine Learning-Project
Spark-lin
12/3/2016
Pratical Machine Learning-Final Project
1-Project Introduction
1.1-Background
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, your goal will be to use data from accelerometers on the belt, forearm, arm, and dumbell of 6 participants. They were asked to perform barbell lifts correctly and incorrectly in 5 different ways. More information is available from the website here: http://groupware.les.inf.puc-rio.br/har (see the section on the Weight Lifting Exercise Dataset).
1.2-Data Source
The training data for this project are available here:
https://d396qusza40orc.cloudfront.net/predmachlearn/pml-training.csv
The test data are available here:
https://d396qusza40orc.cloudfront.net/predmachlearn/pml-testing.csv
The data for this project come from this source: http://groupware.les.inf.puc-rio.br/har. If you use the document you create for this class for any purpose please cite them as they have been very generous in allowing their data to be used for this kind of assignment.
1.3-Goal of Project
The goal of this course project is to predict the manner in which they did the exercise. This is the “classe” variable in the training set. You may use any of the other variables to predict with.
You should create a report describing how you built your model, how you used cross validation, what you think the expected out of sample error is, and rationalize why you made the choices you did. You will also use your prediction model to predict 20 different test cases.
# Load required packages
library(caret)## Loading required package: lattice## Loading required package: ggplot2library(rpart)
library(rpart.plot)
library(rattle)## Rattle: A free graphical interface for data mining with R.
## Version 4.1.0 Copyright (c) 2006-2015 Togaware Pty Ltd.
## Type 'rattle()' to shake, rattle, and roll your data.library(RColorBrewer)
library(randomForest)## randomForest 4.6-12## Type rfNews() to see new features/changes/bug fixes.##
## Attaching package: 'randomForest'## The following object is masked from 'package:ggplot2':
##
## marginlibrary(knitr)2-Getting and cleaning data
TrainUrl <- "https://d396qusza40orc.cloudfront.net/predmachlearn/pml-training.csv"
TestUrl <- "https://d396qusza40orc.cloudfront.net/predmachlearn/pml-testing.csv"
# 2.1-Load and read data
TrainData <- read.csv(url(TrainUrl), na.strings = 'NA')
TestData <- read.csv(url(TestUrl), na.strings = 'NA')
# 2.2-Create a Data partition with the Trainset (only the training data)
inTrain <- createDataPartition(TrainData$classe, p = 0.7, list = FALSE)
TrainSet <- TrainData[inTrain, ]
TestSet <- TrainData[-inTrain, ] # don't put wrongly TestData
dim(TrainSet)## [1] 13737 160dim(TestSet)## [1] 5885 160Cleaning data
# Viewing training data set
names(TrainSet)## [1] "X" "user_name"
## [3] "raw_timestamp_part_1" "raw_timestamp_part_2"
## [5] "cvtd_timestamp" "new_window"
## [7] "num_window" "roll_belt"
## [9] "pitch_belt" "yaw_belt"
## [11] "total_accel_belt" "kurtosis_roll_belt"
## [13] "kurtosis_picth_belt" "kurtosis_yaw_belt"
## [15] "skewness_roll_belt" "skewness_roll_belt.1"
## [17] "skewness_yaw_belt" "max_roll_belt"
## [19] "max_picth_belt" "max_yaw_belt"
## [21] "min_roll_belt" "min_pitch_belt"
## [23] "min_yaw_belt" "amplitude_roll_belt"
## [25] "amplitude_pitch_belt" "amplitude_yaw_belt"
## [27] "var_total_accel_belt" "avg_roll_belt"
## [29] "stddev_roll_belt" "var_roll_belt"
## [31] "avg_pitch_belt" "stddev_pitch_belt"
## [33] "var_pitch_belt" "avg_yaw_belt"
## [35] "stddev_yaw_belt" "var_yaw_belt"
## [37] "gyros_belt_x" "gyros_belt_y"
## [39] "gyros_belt_z" "accel_belt_x"
## [41] "accel_belt_y" "accel_belt_z"
## [43] "magnet_belt_x" "magnet_belt_y"
## [45] "magnet_belt_z" "roll_arm"
## [47] "pitch_arm" "yaw_arm"
## [49] "total_accel_arm" "var_accel_arm"
## [51] "avg_roll_arm" "stddev_roll_arm"
## [53] "var_roll_arm" "avg_pitch_arm"
## [55] "stddev_pitch_arm" "var_pitch_arm"
## [57] "avg_yaw_arm" "stddev_yaw_arm"
## [59] "var_yaw_arm" "gyros_arm_x"
## [61] "gyros_arm_y" "gyros_arm_z"
## [63] "accel_arm_x" "accel_arm_y"
## [65] "accel_arm_z" "magnet_arm_x"
## [67] "magnet_arm_y" "magnet_arm_z"
## [69] "kurtosis_roll_arm" "kurtosis_picth_arm"
## [71] "kurtosis_yaw_arm" "skewness_roll_arm"
## [73] "skewness_pitch_arm" "skewness_yaw_arm"
## [75] "max_roll_arm" "max_picth_arm"
## [77] "max_yaw_arm" "min_roll_arm"
## [79] "min_pitch_arm" "min_yaw_arm"
## [81] "amplitude_roll_arm" "amplitude_pitch_arm"
## [83] "amplitude_yaw_arm" "roll_dumbbell"
## [85] "pitch_dumbbell" "yaw_dumbbell"
## [87] "kurtosis_roll_dumbbell" "kurtosis_picth_dumbbell"
## [89] "kurtosis_yaw_dumbbell" "skewness_roll_dumbbell"
## [91] "skewness_pitch_dumbbell" "skewness_yaw_dumbbell"
## [93] "max_roll_dumbbell" "max_picth_dumbbell"
## [95] "max_yaw_dumbbell" "min_roll_dumbbell"
## [97] "min_pitch_dumbbell" "min_yaw_dumbbell"
## [99] "amplitude_roll_dumbbell" "amplitude_pitch_dumbbell"
## [101] "amplitude_yaw_dumbbell" "total_accel_dumbbell"
## [103] "var_accel_dumbbell" "avg_roll_dumbbell"
## [105] "stddev_roll_dumbbell" "var_roll_dumbbell"
## [107] "avg_pitch_dumbbell" "stddev_pitch_dumbbell"
## [109] "var_pitch_dumbbell" "avg_yaw_dumbbell"
## [111] "stddev_yaw_dumbbell" "var_yaw_dumbbell"
## [113] "gyros_dumbbell_x" "gyros_dumbbell_y"
## [115] "gyros_dumbbell_z" "accel_dumbbell_x"
## [117] "accel_dumbbell_y" "accel_dumbbell_z"
## [119] "magnet_dumbbell_x" "magnet_dumbbell_y"
## [121] "magnet_dumbbell_z" "roll_forearm"
## [123] "pitch_forearm" "yaw_forearm"
## [125] "kurtosis_roll_forearm" "kurtosis_picth_forearm"
## [127] "kurtosis_yaw_forearm" "skewness_roll_forearm"
## [129] "skewness_pitch_forearm" "skewness_yaw_forearm"
## [131] "max_roll_forearm" "max_picth_forearm"
## [133] "max_yaw_forearm" "min_roll_forearm"
## [135] "min_pitch_forearm" "min_yaw_forearm"
## [137] "amplitude_roll_forearm" "amplitude_pitch_forearm"
## [139] "amplitude_yaw_forearm" "total_accel_forearm"
## [141] "var_accel_forearm" "avg_roll_forearm"
## [143] "stddev_roll_forearm" "var_roll_forearm"
## [145] "avg_pitch_forearm" "stddev_pitch_forearm"
## [147] "var_pitch_forearm" "avg_yaw_forearm"
## [149] "stddev_yaw_forearm" "var_yaw_forearm"
## [151] "gyros_forearm_x" "gyros_forearm_y"
## [153] "gyros_forearm_z" "accel_forearm_x"
## [155] "accel_forearm_y" "accel_forearm_z"
## [157] "magnet_forearm_x" "magnet_forearm_y"
## [159] "magnet_forearm_z" "classe"head(TrainSet)## X user_name raw_timestamp_part_1 raw_timestamp_part_2 cvtd_timestamp
## 1 1 carlitos 1323084231 788290 05/12/2011 11:23
## 2 2 carlitos 1323084231 808298 05/12/2011 11:23
## 3 3 carlitos 1323084231 820366 05/12/2011 11:23
## 4 4 carlitos 1323084232 120339 05/12/2011 11:23
## 7 7 carlitos 1323084232 368296 05/12/2011 11:23
## 11 11 carlitos 1323084232 500302 05/12/2011 11:23
## new_window num_window roll_belt pitch_belt yaw_belt total_accel_belt
## 1 no 11 1.41 8.07 -94.4 3
## 2 no 11 1.41 8.07 -94.4 3
## 3 no 11 1.42 8.07 -94.4 3
## 4 no 12 1.48 8.05 -94.4 3
## 7 no 12 1.42 8.09 -94.4 3
## 11 no 12 1.45 8.18 -94.4 3
## kurtosis_roll_belt kurtosis_picth_belt kurtosis_yaw_belt
## 1
## 2
## 3
## 4
## 7
## 11
## skewness_roll_belt skewness_roll_belt.1 skewness_yaw_belt max_roll_belt
## 1 NA
## 2 NA
## 3 NA
## 4 NA
## 7 NA
## 11 NA
## max_picth_belt max_yaw_belt min_roll_belt min_pitch_belt min_yaw_belt
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 7 NA NA NA
## 11 NA NA NA
## amplitude_roll_belt amplitude_pitch_belt amplitude_yaw_belt
## 1 NA NA
## 2 NA NA
## 3 NA NA
## 4 NA NA
## 7 NA NA
## 11 NA NA
## var_total_accel_belt avg_roll_belt stddev_roll_belt var_roll_belt
## 1 NA NA NA NA
## 2 NA NA NA NA
## 3 NA NA NA NA
## 4 NA NA NA NA
## 7 NA NA NA NA
## 11 NA NA NA NA
## avg_pitch_belt stddev_pitch_belt var_pitch_belt avg_yaw_belt
## 1 NA NA NA NA
## 2 NA NA NA NA
## 3 NA NA NA NA
## 4 NA NA NA NA
## 7 NA NA NA NA
## 11 NA NA NA NA
## stddev_yaw_belt var_yaw_belt gyros_belt_x gyros_belt_y gyros_belt_z
## 1 NA NA 0.00 0 -0.02
## 2 NA NA 0.02 0 -0.02
## 3 NA NA 0.00 0 -0.02
## 4 NA NA 0.02 0 -0.03
## 7 NA NA 0.02 0 -0.02
## 11 NA NA 0.03 0 -0.02
## accel_belt_x accel_belt_y accel_belt_z magnet_belt_x magnet_belt_y
## 1 -21 4 22 -3 599
## 2 -22 4 22 -7 608
## 3 -20 5 23 -2 600
## 4 -22 3 21 -6 604
## 7 -22 3 21 -4 599
## 11 -21 2 23 -5 596
## magnet_belt_z roll_arm pitch_arm yaw_arm total_accel_arm var_accel_arm
## 1 -313 -128 22.5 -161 34 NA
## 2 -311 -128 22.5 -161 34 NA
## 3 -305 -128 22.5 -161 34 NA
## 4 -310 -128 22.1 -161 34 NA
## 7 -311 -128 21.9 -161 34 NA
## 11 -317 -128 21.5 -161 34 NA
## avg_roll_arm stddev_roll_arm var_roll_arm avg_pitch_arm
## 1 NA NA NA NA
## 2 NA NA NA NA
## 3 NA NA NA NA
## 4 NA NA NA NA
## 7 NA NA NA NA
## 11 NA NA NA NA
## stddev_pitch_arm var_pitch_arm avg_yaw_arm stddev_yaw_arm var_yaw_arm
## 1 NA NA NA NA NA
## 2 NA NA NA NA NA
## 3 NA NA NA NA NA
## 4 NA NA NA NA NA
## 7 NA NA NA NA NA
## 11 NA NA NA NA NA
## gyros_arm_x gyros_arm_y gyros_arm_z accel_arm_x accel_arm_y accel_arm_z
## 1 0.00 0.00 -0.02 -288 109 -123
## 2 0.02 -0.02 -0.02 -290 110 -125
## 3 0.02 -0.02 -0.02 -289 110 -126
## 4 0.02 -0.03 0.02 -289 111 -123
## 7 0.00 -0.03 0.00 -289 111 -125
## 11 0.02 -0.03 0.00 -290 110 -123
## magnet_arm_x magnet_arm_y magnet_arm_z kurtosis_roll_arm
## 1 -368 337 516
## 2 -369 337 513
## 3 -368 344 513
## 4 -372 344 512
## 7 -373 336 509
## 11 -366 339 509
## kurtosis_picth_arm kurtosis_yaw_arm skewness_roll_arm
## 1
## 2
## 3
## 4
## 7
## 11
## skewness_pitch_arm skewness_yaw_arm max_roll_arm max_picth_arm
## 1 NA NA
## 2 NA NA
## 3 NA NA
## 4 NA NA
## 7 NA NA
## 11 NA NA
## max_yaw_arm min_roll_arm min_pitch_arm min_yaw_arm amplitude_roll_arm
## 1 NA NA NA NA NA
## 2 NA NA NA NA NA
## 3 NA NA NA NA NA
## 4 NA NA NA NA NA
## 7 NA NA NA NA NA
## 11 NA NA NA NA NA
## amplitude_pitch_arm amplitude_yaw_arm roll_dumbbell pitch_dumbbell
## 1 NA NA 13.05217 -70.49400
## 2 NA NA 13.13074 -70.63751
## 3 NA NA 12.85075 -70.27812
## 4 NA NA 13.43120 -70.39379
## 7 NA NA 13.12695 -70.24757
## 11 NA NA 13.13074 -70.63751
## yaw_dumbbell kurtosis_roll_dumbbell kurtosis_picth_dumbbell
## 1 -84.87394
## 2 -84.71065
## 3 -85.14078
## 4 -84.87363
## 7 -85.09961
## 11 -84.71065
## kurtosis_yaw_dumbbell skewness_roll_dumbbell skewness_pitch_dumbbell
## 1
## 2
## 3
## 4
## 7
## 11
## skewness_yaw_dumbbell max_roll_dumbbell max_picth_dumbbell
## 1 NA NA
## 2 NA NA
## 3 NA NA
## 4 NA NA
## 7 NA NA
## 11 NA NA
## max_yaw_dumbbell min_roll_dumbbell min_pitch_dumbbell min_yaw_dumbbell
## 1 NA NA
## 2 NA NA
## 3 NA NA
## 4 NA NA
## 7 NA NA
## 11 NA NA
## amplitude_roll_dumbbell amplitude_pitch_dumbbell amplitude_yaw_dumbbell
## 1 NA NA
## 2 NA NA
## 3 NA NA
## 4 NA NA
## 7 NA NA
## 11 NA NA
## total_accel_dumbbell var_accel_dumbbell avg_roll_dumbbell
## 1 37 NA NA
## 2 37 NA NA
## 3 37 NA NA
## 4 37 NA NA
## 7 37 NA NA
## 11 37 NA NA
## stddev_roll_dumbbell var_roll_dumbbell avg_pitch_dumbbell
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 7 NA NA NA
## 11 NA NA NA
## stddev_pitch_dumbbell var_pitch_dumbbell avg_yaw_dumbbell
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 7 NA NA NA
## 11 NA NA NA
## stddev_yaw_dumbbell var_yaw_dumbbell gyros_dumbbell_x gyros_dumbbell_y
## 1 NA NA 0 -0.02
## 2 NA NA 0 -0.02
## 3 NA NA 0 -0.02
## 4 NA NA 0 -0.02
## 7 NA NA 0 -0.02
## 11 NA NA 0 -0.02
## gyros_dumbbell_z accel_dumbbell_x accel_dumbbell_y accel_dumbbell_z
## 1 0.00 -234 47 -271
## 2 0.00 -233 47 -269
## 3 0.00 -232 46 -270
## 4 -0.02 -232 48 -269
## 7 0.00 -232 47 -270
## 11 0.00 -233 47 -269
## magnet_dumbbell_x magnet_dumbbell_y magnet_dumbbell_z roll_forearm
## 1 -559 293 -65 28.4
## 2 -555 296 -64 28.3
## 3 -561 298 -63 28.3
## 4 -552 303 -60 28.1
## 7 -551 295 -70 27.9
## 11 -564 299 -64 27.6
## pitch_forearm yaw_forearm kurtosis_roll_forearm kurtosis_picth_forearm
## 1 -63.9 -153
## 2 -63.9 -153
## 3 -63.9 -152
## 4 -63.9 -152
## 7 -63.9 -152
## 11 -63.8 -152
## kurtosis_yaw_forearm skewness_roll_forearm skewness_pitch_forearm
## 1
## 2
## 3
## 4
## 7
## 11
## skewness_yaw_forearm max_roll_forearm max_picth_forearm max_yaw_forearm
## 1 NA NA
## 2 NA NA
## 3 NA NA
## 4 NA NA
## 7 NA NA
## 11 NA NA
## min_roll_forearm min_pitch_forearm min_yaw_forearm
## 1 NA NA
## 2 NA NA
## 3 NA NA
## 4 NA NA
## 7 NA NA
## 11 NA NA
## amplitude_roll_forearm amplitude_pitch_forearm amplitude_yaw_forearm
## 1 NA NA
## 2 NA NA
## 3 NA NA
## 4 NA NA
## 7 NA NA
## 11 NA NA
## total_accel_forearm var_accel_forearm avg_roll_forearm
## 1 36 NA NA
## 2 36 NA NA
## 3 36 NA NA
## 4 36 NA NA
## 7 36 NA NA
## 11 36 NA NA
## stddev_roll_forearm var_roll_forearm avg_pitch_forearm
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 7 NA NA NA
## 11 NA NA NA
## stddev_pitch_forearm var_pitch_forearm avg_yaw_forearm
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 7 NA NA NA
## 11 NA NA NA
## stddev_yaw_forearm var_yaw_forearm gyros_forearm_x gyros_forearm_y
## 1 NA NA 0.03 0.00
## 2 NA NA 0.02 0.00
## 3 NA NA 0.03 -0.02
## 4 NA NA 0.02 -0.02
## 7 NA NA 0.02 0.00
## 11 NA NA 0.02 -0.02
## gyros_forearm_z accel_forearm_x accel_forearm_y accel_forearm_z
## 1 -0.02 192 203 -215
## 2 -0.02 192 203 -216
## 3 0.00 196 204 -213
## 4 0.00 189 206 -214
## 7 -0.02 195 205 -215
## 11 -0.02 193 205 -214
## magnet_forearm_x magnet_forearm_y magnet_forearm_z classe
## 1 -17 654 476 A
## 2 -18 661 473 A
## 3 -18 658 469 A
## 4 -16 658 469 A
## 7 -18 659 470 A
## 11 -17 657 465 AWe observe that there are variables which are irrelavant for our analysis. Such as “X” and “user_variable” other NA value, Near Zero Variance (NZV) variables and ID variables
TrainSet <- TrainSet[, -(1:2)]
TestSet <- TestSet[, -(1:2)]
# 2.3-Remove Near Zero variable Variance
NZV <- nearZeroVar(TrainSet)
TrainSet <- TrainSet[, -NZV]
TestSet <- TestSet[, -NZV]
dim(TrainSet)## [1] 13737 103dim(TestSet )## [1] 5885 103# 2.4-Remove variable that have a possibility more than 70& of value of "NA"
NA_value <- sapply(TrainSet, function(x) mean(is.na(x)))
TrainSet <- TrainSet[, (NA_value > 0.7) == FALSE]
TestSet <- TestSet[, (NA_value > 0.7) == FALSE]
dim(TrainSet)## [1] 13737 57dim(TestSet)## [1] 5885 57# 2.5-Remove the ID variables
TrainSet <- TrainSet[, -(1:5)]
TestSet <- TestSet[, -(1:5)]
dim(TrainSet)## [1] 13737 52dim(TestSet)## [1] 5885 52After cleaning the irrelevant variables, we manage to reduce the number of variables for analysis to 52
3-Correaltion among the variables
Let’s run a correlation analysis among the variables to see if some of varaibles are highly correlated to each other.
library(corrplot)
corelationMatrix <- cor(TrainSet[, -52])
corrplot(corelationMatrix, order = "hclust", method = "color", type = "lower", tl.cex = 0.7, tl.col = rgb(0, 0, 0))
# order = "FPC", refers to the first principal component order
# order = "hclust", refers to hierarchical clustering order.
# type should only be one of “full”, “lower”, “upper” value.4-Prediction Model
In this part, we will use 3 prediction model to simulate the regression in the TrainSet,contrast related results and accuracy level, then select the highest accuracy methode for the quiz prediction.
The prediction methode used are: - Decision Trees - Random Forests - Generalized Boosted Regression
4.1- Decision Trees Methode
set.seed(12000)
modelFit_DecisionTree <- rpart(classe ~ ., data = TrainSet, method = "class")
fancyRpartPlot(modelFit_DecisionTree)## Warning: labs do not fit even at cex 0.15, there may be some overplotting
# Decision Tree Prediction on Test dataset
Pred_DecisionTree <- predict(modelFit_DecisionTree, newdata = TestSet, type = "class")
ConfusionMatrix_DecisionTree <- confusionMatrix(Pred_DecisionTree, TestSet$classe)
ConfusionMatrix_DecisionTree## Confusion Matrix and Statistics
##
## Reference
## Prediction A B C D E
## A 1460 185 19 36 51
## B 62 647 59 80 156
## C 26 79 779 137 155
## D 106 177 143 654 128
## E 20 51 26 57 592
##
## Overall Statistics
##
## Accuracy : 0.7021
## 95% CI : (0.6903, 0.7138)
## No Information Rate : 0.2845
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.6232
## Mcnemar's Test P-Value : < 2.2e-16
##
## Statistics by Class:
##
## Class: A Class: B Class: C Class: D Class: E
## Sensitivity 0.8722 0.5680 0.7593 0.6784 0.5471
## Specificity 0.9309 0.9248 0.9183 0.8874 0.9679
## Pos Pred Value 0.8338 0.6444 0.6624 0.5414 0.7936
## Neg Pred Value 0.9482 0.8992 0.9475 0.9337 0.9047
## Prevalence 0.2845 0.1935 0.1743 0.1638 0.1839
## Detection Rate 0.2481 0.1099 0.1324 0.1111 0.1006
## Detection Prevalence 0.2975 0.1706 0.1998 0.2053 0.1268
## Balanced Accuracy 0.9015 0.7464 0.8388 0.7829 0.7575# Plot Decision Tree Confusion Matrix
plot(ConfusionMatrix_DecisionTree$table, col = ConfusionMatrix_DecisionTree$byClass,
main = paste("Decision Tree Confusion Matrix: Accuracy =",
round(ConfusionMatrix_DecisionTree$overall['Accuracy'], 4)))
###### 4.2- Random Forests Methode
# Random Forests Prediction on Test dataset
set.seed(12000)
ModelFit_RanForest <- randomForest(classe~., data = TrainSet)
Pred_RanForest <- predict(ModelFit_RanForest, TestSet, type = "class")
ConfusionMatrix_RanForest <- confusionMatrix(Pred_RanForest, TestSet$classe)
ConfusionMatrix_RanForest## Confusion Matrix and Statistics
##
## Reference
## Prediction A B C D E
## A 1674 7 0 0 0
## B 0 1131 4 0 0
## C 0 1 1022 11 2
## D 0 0 0 953 4
## E 0 0 0 0 1076
##
## Overall Statistics
##
## Accuracy : 0.9951
## 95% CI : (0.9929, 0.9967)
## No Information Rate : 0.2845
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.9938
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: A Class: B Class: C Class: D Class: E
## Sensitivity 1.0000 0.9930 0.9961 0.9886 0.9945
## Specificity 0.9983 0.9992 0.9971 0.9992 1.0000
## Pos Pred Value 0.9958 0.9965 0.9865 0.9958 1.0000
## Neg Pred Value 1.0000 0.9983 0.9992 0.9978 0.9988
## Prevalence 0.2845 0.1935 0.1743 0.1638 0.1839
## Detection Rate 0.2845 0.1922 0.1737 0.1619 0.1828
## Detection Prevalence 0.2856 0.1929 0.1760 0.1626 0.1828
## Balanced Accuracy 0.9992 0.9961 0.9966 0.9939 0.9972# Plot ModelFit Random Forest
plot(ModelFit_RanForest)
# Plot Random Forest Confusion Matrix
plot(ConfusionMatrix_RanForest$table, col = ConfusionMatrix_RanForest$byClass,
main = paste("Random Forest Confusion Matrix: Accuracy =",
round(ConfusionMatrix_RanForest$overall['Accuracy'], 4)))
###### 4.3-Generalized Boosted Regression Methode (GBM)
ModelFit_Control <- trainControl(method = "repeatedcv", number = 5, repeats = 1)
ModelFit_GBM <- train(classe ~ ., data = TrainSet,
method = "gbm",
trControl = ModelFit_Control,
verbose = FALSE)## Loading required package: gbm## Loading required package: survival##
## Attaching package: 'survival'## The following object is masked from 'package:caret':
##
## cluster## Loading required package: splines## Loading required package: parallel## Loaded gbm 2.1.1## Loading required package: plyrGBM_FinalModel <- ModelFit_GBM$finalModel
GBM_FinalModel## A gradient boosted model with multinomial loss function.
## 150 iterations were performed.
## There were 51 predictors of which 43 had non-zero influence.# Generalized Boosted Regression Methode (GBM) Prediction on Test dataset
Pred_GBM <- predict(ModelFit_GBM, newdata = TestSet)
ConfusionMatrix_GBM <- confusionMatrix(Pred_GBM, TestSet$classe)
ConfusionMatrix_GBM## Confusion Matrix and Statistics
##
## Reference
## Prediction A B C D E
## A 1645 47 0 1 0
## B 20 1056 40 3 11
## C 7 29 972 35 11
## D 2 1 12 917 17
## E 0 6 2 8 1043
##
## Overall Statistics
##
## Accuracy : 0.9572
## 95% CI : (0.9517, 0.9622)
## No Information Rate : 0.2845
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.9458
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: A Class: B Class: C Class: D Class: E
## Sensitivity 0.9827 0.9271 0.9474 0.9512 0.9640
## Specificity 0.9886 0.9844 0.9831 0.9935 0.9967
## Pos Pred Value 0.9716 0.9345 0.9222 0.9663 0.9849
## Neg Pred Value 0.9931 0.9825 0.9888 0.9905 0.9919
## Prevalence 0.2845 0.1935 0.1743 0.1638 0.1839
## Detection Rate 0.2795 0.1794 0.1652 0.1558 0.1772
## Detection Prevalence 0.2877 0.1920 0.1791 0.1613 0.1799
## Balanced Accuracy 0.9856 0.9558 0.9652 0.9724 0.9803# Plot Generalized Boosted Regression Methode (GBM) Confusion Matrix
plot(ConfusionMatrix_GBM$table, col = ConfusionMatrix_GBM$byClass,
main = paste("GBM Confusion Matrix: Accuracy =",
round(ConfusionMatrix_GBM$overall['Accuracy'], 4)))
###### 4.4-Contrast the acuracy of the Three Prediction Methodes
# Comparison of the Three Prediction Methodes Accuracy result
print(paste("Decision Tree Confusion Matrix: Accuracy =",
round(ConfusionMatrix_DecisionTree$overall['Accuracy'], 4)))## [1] "Decision Tree Confusion Matrix: Accuracy = 0.7021"print(paste("Random Forest Confusion Matrix: Accuracy =",
round(ConfusionMatrix_RanForest$overall['Accuracy'], 4)))## [1] "Random Forest Confusion Matrix: Accuracy = 0.9951"print(paste("GBM Confusion Matrix: Accuracy: Accuracy =",
round(ConfusionMatrix_GBM$overall['Accuracy'], 4)))## [1] "GBM Confusion Matrix: Accuracy: Accuracy = 0.9572"As the we can see from the above Accuracy Comparison, the Random Forest model have the highest level of accuracy of 0.9949. The expected out of-sample error = 100% - 99.49% = 0.51%.
Therefore, we would choose the Random Forest model to precit 20 different test cases.
5-Apply the the Random Forest model to precit 20 different test cases.(TestData)
The 20 Quiz tested results are shown below:
Show the Extract of the first 20 results (Total more than 8000 entries)
# Show the Extract of the first 20 results (Total morethan 8000 entries)
Pred_Test[1:20]## 5 6 8 9 10 18 20 27 29 32 34 39 43 44 50 54 56 57 58 59
## A A A A A A A A A A A A A A A A A A A A
## Levels: A B C D E