Hands-on Ensemble Learning with R

Cover
Copyright
Contributors
Table of Contents
Preface
Chapter 1: Introduction to Ensemble Techniques
	Datasets
		Hypothyroid
		Waveform
		German Credit
		Iris
		Pima Indians Diabetes
		US Crime
		Overseas visitors
		Primary Biliary Cirrhosis
		Multishapes
		Board Stiffness
	Statistical/machine learning models
		Logistic regression model
			Logistic regression for hypothyroid classification
		Neural networks
			Neural network for hypothyroid classification
		Naïve Bayes classifier
			Naïve Bayes for hypothyroid classification
		Decision tree
			Decision tree for hypothyroid classification
		Support vector machines
			SVM for hypothyroid classification
	The right model dilemma!
	An ensemble purview
	Complementary statistical tests
		Permutation test
		Chi-square and McNemar test
		ROC test
	Summary
Chapter 2: Bootstrapping
	Technical requirements
	The jackknife technique
		The jackknife method for mean and variance
		Pseudovalues method for survival data
	Bootstrap – a statistical method
		The standard error of correlation coefficient
		The parametric bootstrap
		Eigen values
			Rule of thumb
	The boot package
	Bootstrap and testing hypotheses
	Bootstrapping regression models
	Bootstrapping survival models*
	Bootstrapping time series models*
	Summary
Chapter 3: Bagging
	Technical requirements
	Classification trees and pruning
	Bagging
	k-NN classifier
		Analyzing waveform data
	k-NN bagging
	Summary
Chapter 4: Random Forests
	Technical requirements
	Random Forests
	Variable importance
	Proximity plots
	Random Forest nuances
	Comparisons with bagging
	Missing data imputation
	Clustering with Random Forest
	Summary
Chapter 5: The Bare Bones Boosting Algorithms
	Technical requirements
	The general boosting algorithm
	Adaptive boosting
	Gradient boosting
		Building it from scratch
		Squared-error loss function
	Using the adabag and gbm packages
	Variable importance
	Comparing bagging, random forests, and boosting
	Summary
Chapter 6: Boosting Refinements
	Technical requirements
	Why does boosting work?
	The gbm package
		Boosting for count data
		Boosting for survival data
	The xgboost package
	The h2o package
	Summary
Chapter 7: The General Ensemble Technique
	Technical requirements
	Why does ensembling work?
	Ensembling by voting
		Majority voting
		Weighted voting
	Ensembling by averaging
		Simple averaging
		Weight averaging
	Stack ensembling
	Summary
Chapter 8: Ensemble Diagnostics
	Technical requirements
	What is ensemble diagnostics?
	Ensemble diversity
		Numeric prediction
		Class prediction
	Pairwise measure
		Disagreement measure
		Yule\'s or Q-statistic
		Correlation coefficient measure
		Cohen\'s statistic
		Double-fault measure
	Interrating agreement
		Entropy measure
		Kohavi-Wolpert measure
		Disagreement measure for ensemble
		Measurement of interrater agreement
	Summary
Chapter 9: Ensembling Regression Models
	Technical requirements
	Pre-processing the housing data
	Visualization and variable reduction
		Variable clustering
	Regression models
		Linear regression model
		Neural networks
		Regression tree
		Prediction for regression models
	Bagging and Random Forests
	Boosting regression models
	Stacking methods for regression models
	Summary
Chapter 10: Ensembling Survival Models
	Core concepts of survival analysis
	Nonparametric inference
	Regression models – parametric and Cox proportional hazards models
	Survival tree
	Ensemble survival models
	Summary
Chapter 11: Ensembling Time Series Models
	Technical requirements
	Time series datasets
		AirPassengers
		co2
		uspop
		gas
		Car Sales
		austres
		WWWusage
	Time series visualization
	Core concepts and metrics
	Essential time series models
		Naïve forecasting
		Seasonal, trend, and loess fitting
		Exponential smoothing state space model
		Auto-regressive Integrated Moving Average (ARIMA) models
		Auto-regressive neural networks
		Messing it all up
	Bagging and time series
	Ensemble time series models
	Summary
Chapter 12: What\'s Next?
Bibliography
	References
	R package references
Other Books You May Enjoy
Index