Book Image

Machine Learning Algorithms

Book Image

Machine Learning Algorithms

Overview of this book

In this book, you will learn all the important machine learning algorithms that are commonly used in the field of data science. These algorithms can be used for supervised as well as unsupervised learning, reinforcement learning, and semi-supervised learning. The algorithms that are covered in this book are linear regression, logistic regression, SVM, naïve Bayes, k-means, random forest, TensorFlow and feature engineering. In this book, you will how to use these algorithms to resolve your problems, and how they work. This book will also introduce you to natural language processing and recommendation systems, which help you to run multiple algorithms simultaneously. On completion of the book, you will know how to pick the right machine learning algorithm for clustering, classification, or regression for your problem

Related Content you might be interested in

Current Title:

Small book image
Machine Learning Algorithms
Jul, 2017 | 360 pages
No titles found
Table of Contents (22 chapters)
Title Page
Title Page
Credits
Credits
About the Author
About the Author
About the Reviewers
About the Reviewers
www.PacktPub.com
www.PacktPub.com
Customer Feedback
Customer Feedback
Preface
Preface
Free Chapter
1
A Gentle Introduction to Machine Learning
A Gentle Introduction to Machine Learning
Introduction - classic and adaptive machines
Only learning matters
Beyond machine learning - deep learning and bio-inspired adaptive systems
Machine learning and big data
Further reading
Summary
2
Important Elements in Machine Learning
Important Elements in Machine Learning
Data formats
Learnability
Statistical learning approaches
Elements of information theory
References
Summary
3
Feature Selection and Feature Engineering
Feature Selection and Feature Engineering
scikit-learn toy datasets
Creating training and test sets
Managing categorical data
Managing missing features
Data scaling and normalization
Feature selection and filtering
Principal component analysis
Atom extraction and dictionary learning
References
Summary
4
Linear Regression
Linear Regression
Linear models
A bidimensional example
Linear regression with scikit-learn and higher dimensionality
Ridge, Lasso, and ElasticNet
Robust regression with random sample consensus
Polynomial regression
Isotonic regression
References
Summary
5
Logistic Regression
Logistic Regression
Linear classification
Logistic regression
Implementation and optimizations
Stochastic gradient descent algorithms
Finding the optimal hyperparameters through grid search
Classification metrics
ROC curve
Summary
6
Naive Bayes
Naive Bayes
Bayes' theorem
Naive Bayes classifiers
Naive Bayes in scikit-learn
References
Summary
7
Support Vector Machines
Support Vector Machines
Linear support vector machines
scikit-learn implementation
Controlled support vector machines
Support vector regression
References
Summary
8
Decision Trees and Ensemble Learning
Decision Trees and Ensemble Learning
Binary decision trees
Decision tree classification with scikit-learn
Ensemble learning
References
Summary
9
Clustering Fundamentals
Clustering Fundamentals
Clustering basics
Evaluation methods based on the ground truth
References
Summary
10
Hierarchical Clustering
Hierarchical Clustering
Hierarchical strategies
Agglomerative clustering
References
Summary
11
Introduction to Recommendation Systems
Introduction to Recommendation Systems
Naive user-based systems
Content-based systems
Model-free (or memory-based) collaborative filtering
Model-based collaborative filtering
References
Summary
12
Introduction to Natural Language Processing
Introduction to Natural Language Processing
NLTK and built-in corpora
The bag-of-words strategy
A sample text classifier based on the Reuters corpus
References
Summary
13
Topic Modeling and Sentiment Analysis in NLP
Topic Modeling and Sentiment Analysis in NLP
Topic modeling
Sentiment analysis
References
Summary
14
A Brief Introduction to Deep Learning and TensorFlow
A Brief Introduction to Deep Learning and TensorFlow
Deep learning at a glance
A brief introduction to TensorFlow
A quick glimpse inside Keras
References
Summary
15
Creating a Machine Learning Architecture
Creating a Machine Learning Architecture
Machine learning architectures
scikit-learn tools for machine learning architectures
References
Summary

Elements of information theory

A machine learning problem can also be analyzed in terms of information transfer or exchange. Our dataset is composed of n features, which are considered independent (for simplicity, even if it's often a realistic assumption) drawn from n different statistical distributions. Therefore, there are n probability density functions pi(x) which must be approximated through other nqi(x) functions. In any machine learning task, it's very important to understand how two corresponding distributions diverge and what is the amount of information we lose when approximating the original dataset.

The most useful measure is called entropy:

This value is proportional to the uncertainty of X and it's measured in bits (if the logarithm has another base, this unit can change too). For many purposes, a high entropy is preferable, because it means that a certain feature contains more information. For example, in tossing a coin (two possible outcomes), H(X) = 1 bit, but if the number...

Previous Section
End of Section 5
Next Section