Machine Learning · the Art and Science of Algorithms That Make Sense of Data by Flach, Peter -- Read -- Imperial Library of Trantor

Index

Cover Page Title Page Copyright Page Dedication Page Contents Preface Prologue: A machine Learning Sampler 1 - The Ingredients of Machine Learning

1.1 Tasks: The Problems that can be Solved with Machine Learning

Looking for Structure Evaluating Performance on a Task

1.2 Models: The Output of Machine Learning

Geometric Models Probabilistic Models Logical Models Grouping and Grading

1.3 Features: The Workhorses of Machine Learning

Two Uses of Features Feature Construction and Transformation Interaction Between Features

1.4 Summary and outlook

What you’ll find in the Rest of the Book

2 - Binary Classification and Related Tasks

2.1 Classification

Assessing Classification Performance Visualising Classification Performance

2.2 Scoring and Ranking

Assessing and Visualising Ranking Performance Turning Rankers into Classifiers

2.3 Class Probability Estimation

Assessing Class Probability Estimates Turning Rankers into Class Probability Estimators

2.4 Binary Classification and Related tasks: Summary and Further Reading

3 - Beyond Binary Classification

3.1 Handling more than two Classes

Multi-class Classification Multi-class Scores and Probabilities

3.2 Regression 3.3 Unsupervised and Descriptive Learning

Predictive and Descriptive Clustering Other Descriptive Models

3.4 Beyond Binary Classification: Summary and Further Reading

4 - Concept Learning

4.1 The Hypothesis Space

Least General Generalisation Internal Disjunction

4.2 Paths Through the Hypothesis Space

Most General Consistent Hypotheses Closed Concepts

4.3 Beyond Conjunctive Concepts

Using First-order Logic

4.4 Learnability 4.5 Concept Learning: Summary and Further Reading

5 - Tree Models

5.1 Decision Trees 5.2 Ranking and Probability Estimation trees

Sensitivity to Skewed Class Distributions

5.3 Tree Learning as Variance Reduction

Regression Trees Clustering Trees

5.4 Tree models: Summary and Further Reading

6 - Rule Models

6.1 Learning Ordered Rule Lists

Rule lists for Ranking and Probability Estimation

6.2 Learning Unordered Rule Sets

Rule Sets for Ranking and Probability Estimation A Closer Look at Rule Overlap

6.3 Descriptive Rule Learning

Rule Learning for Subgroup Discovery Association Rule Mining

6.4 First-orderrule Learning 6.5 Rule Models: Summary and Further Reading

7 - Linear Models

7.1 The Least-squares Method

Multivariate Linear Regression Regularised Regression Using Least-squares Regression for Classification

7.2 The Perceptron 7.3 Support Vector Machines

Soft Margin Svm

7.4 Obtaining Probabilities from Linear Classifiers 7.5 Going Beyond Linearity with Kernel Methods 7.6 Linear Models: Summary and Further Reading

8 - Distance-based Models

8.1 So many Roads 8.2 Neighbours and Exemplars 8.3 Nearest-neighbour Classification 8.4 Distance-based Clustering

K-means Algorithm Clustering Around Medoids Silhouettes

8.5 Hierarchical Clustering 8.6 From Kernels to Distances 8.7 Distance-based Models: Summary and Further Reading

9 - Probabilistic Models

9.1 The Normal Distribution and its Geometric Interpretations 9.2 Probabilistic Models for Categorical Data

Using a Naive Bayes Model for Classification Training a Naive Bayes model

9.3 Discriminative Learning by Optimising Conditional Likelihood 9.4 Probabilistic Models with Hidden Variables

Expectation-Maximisation Gaussian Mixture Models

9.5 Compression-based Models 9.6 Probabilistic Models: Summary and Further Reading

10 - Features

10.1 Kinds of Feature

Calculations on Features Categorical, Ordinal and Quantitative Features Structured Features

10.2 Feature Transformations

Thresholding and Discretisation Normalisation and Calibration Incomplete Features

10.3 Feature Construction and Selection

Matrix Transformations and Decompositions

10.4 Features: Summary and Further Reading

11 - Model Ensembles

11.1 Bagging and Random Forests 11.2 Boosting

Boosted Rule Learning

11.3 Mapping the Ensemble Landscape

Bias, variance and Margins Other Ensemble Methods Meta-learning

11.4 Model Ensembles: Summary and Further Reading

12 - Machine Learning experiments

12.1 What to Measure 12.2 How to Measure it 12.3 How to Interpret it

Interpretation of Results Over Multiple Data Sets

12.4 Machine Learning Experiments: Summary and Further Reading

Epilogue: Where to go from here Important Points to Remember References Index

← Prev
Back
Next →

← Prev
Back
Next →