You're reading from Mastering Machine Learning Algorithms Expert techniques for implementing popular machine learning algorithms, fine-tuning your models, and understanding how they work

Product type Paperback

Published in Jan 2020

Publisher Packt

ISBN-13 9781838820299

Length 798 pages

Edition 2nd Edition

Languages

Python

Tools

Keras

Concepts

Machine Learning

Authors (2):

Giuseppe Bonaccorso

View More author details

Table of Contents (28) Chapters

Preface

1. Machine Learning Model Fundamentals

2. Loss Functions and Regularization FREE CHAPTER

3. Introduction to Semi-Supervised Learning

4. Advanced Semi-Supervised Classification

5. Graph-Based Semi-Supervised Learning

6. Clustering and Unsupervised Models

7. Advanced Clustering and Unsupervised Models

8. Clustering and Unsupervised Models for Marketing

9. Generalized Linear Models and Regression

10. Introduction to Time-Series Analysis

11. Bayesian Networks and Hidden Markov Models

12. The EM Algorithm

13. Component Analysis and Dimensionality Reduction

14. Hebbian Learning

15. Fundamentals of Ensemble Learning

16. Advanced Boosting Algorithms

17. Modeling Neural Networks

18. Optimizing Neural Networks

19. Deep Convolutional Networks

20. Recurrent Neural Networks

21. Autoencoders

22. Introduction to Generative Adversarial Networks

23. Deep Belief Networks

24. Introduction to Reinforcement Learning

25. Advanced Policy Estimation Algorithms

26. Other Books You May Enjoy

27. Index

Spectral clustering

One of the most common problems with K-means and other similar algorithms is the assumption that we only have hyperspherical clusters. In fact, K-means is insensitive to the angle and assigns a label only according to the closest distance between a point and centroids. The resulting geometry is based on hyperspheres where all points share the same condition to be closer to the same centroid. This condition might be acceptable when the dataset is split into blobs that can be easily embedded into a regular geometric structure. However, it fails whenever the sets are not separable using regular shapes. Let's consider, for example, the following bidimensional dataset:

Sinusoidal dataset

As we are going to see in the example, any attempt to separate the upper sinusoid from the lower one using K-means will fail. The reason is obvious: a circle that contains the upper set will also contain part of the (or the whole) lower set. Considering the criterion...