book

Introduction to Machine Learning with R

by Scott V. Burger

March 2018

Beginner to intermediate

223 pages

5h 38m

English

O'Reilly Media, Inc.

Read now

Unlock full access

Who Should Read This Book?Scope of the BookConventions Used in This BookO’Reilly SafariHow to Contact UsAcknowledgments
Algorithms Versus Models: What’s the Difference?A Note on TerminologyModeling LimitationsStatistics and Computation in ModelingData TrainingCross-ValidationWhy Use R?The GoodR and Machine LearningThe BadSummary
Supervised ModelsRegressionTraining and Testing of DataClassificationLogistic RegressionSupervised Clustering MethodsMixed MethodsTree-Based ModelsRandom ForestsNeural NetworksSupport Vector MachinesUnsupervised LearningUnsupervised Clustering MethodsSummary
BiasSampling in RTraining and TestingRoles of Training and Test SetsWhy Make a Test Set?Training and Test Sets: Regression ModelingTraining and Test Sets: Classification ModelingCross-Validationk-Fold Cross-ValidationSummary
Linear RegressionMultivariate RegressionRegularizationPolynomial RegressionGoodness of Fit with Data—The Perils of OverfittingRoot-Mean-Square ErrorModel Simplicity and Goodness of FitLogistic RegressionThe Motivation for ClassificationThe Decision BoundaryThe Sigmoid FunctionBinary ClassificationMulticlass ClassificationLogistic Regression with CaretSummaryLinear RegressionLogistic Regression
Single-Layer Neural NetworksBuilding a Simple Neural Network by Using RMultiple Compute OutputsHidden Compute NodesMultilayer Neural NetworksNeural Networks for RegressionNeural Networks for ClassificationNeural Networks with caretRegressionClassificationSummary
A Simple Tree ModelDeciding How to Split TreesTree Entropy and Information GainPros and Cons of Decision TreesTree OverfittingPruning TreesDecision Trees for RegressionDecision Trees for ClassificationConditional Inference TreesConditional Inference Tree RegressionConditional Inference Tree ClassificationRandom ForestsRandom Forest RegressionRandom Forest ClassificationSummary
Naive Bayes ClassificationBayesian Statistics in a NutshellApplication of Naive BayesPrincipal Component AnalysisLinear Discriminant AnalysisSupport Vector Machinesk-Nearest NeighborsRegression Using kNNClassification Using kNNSummary
The Titanic DatasetData Wranglingcaret UnleashedImputationData Splittingcaret Under the HoodModel TrainingComparing Multiple caret ModelsSummary

Content preview from Introduction to Machine Learning with R

Chapter 5. Neural Networks in a Nutshell

In Chapter 2, we briefly touched on the topic of neural networks in our exploration of the machine learning landscape. A neural network is a set of equations that we use to calculate an outcome. They aren’t so scary if we think of them as a brain made out of computer code. In some cases, this is closer to reality than we should expect from such a cartoony example. Depending on the number of features we have in our data, the neural network almost becomes a “black box.” In principle, we can display the equations that make up a neural network, but at a certain level, the amount of information becomes too cumbersome to intuit easily.

Neural networks are used far and wide in industry largely due to their accuracy. Sometimes, there are trade-offs between having a highly accurate model, but slow computation speeds, however. Therefore, it’s best to try multiple models and use neural networks only if they work for your particular dataset.

Single-Layer Neural Networks

In Chapter 2, we looked at the development of an AND gate. An AND gate follows logic like this:

x1 <- c(0, 0, 1, 1)
x2 <- c(0, 1, 0, 1)
logic <- data.frame(x1, x2)
logic$AND <- as.numeric(x1 & x2)
logic

##   x1 x2 AND
## 1  0  0   0
## 2  0  1   0
## 3  1  0   0
## 4  1  1   1

If you have two 1 inputs (both TRUE), your output is 1 (TRUE). However, if either of them, or both, are 0 (FALSE), your output is also 0 (FALSE). This computation is somewhat similar to our analysis of logistic regression. In ...