Python Machine Learning By Example

Title Page

Python Machine Learning By Example

Easy-to-follow examples that get you up and running with machine learning

Yuxi (Hayden) Liu

 BIRMINGHAM - MUMBAI

Copyright

Credits

About the Author

Yuxi (Hayden) Liu is currently a data scientist working on messaging app optimization at a multinational online media corporation in Toronto, Canada. He is focusing on social graph mining, social personalization, user demographics and interests prediction, spam detection, and recommendation systems. He has worked for a few years as a data scientist at several programmatic advertising companies, where he applied his machine learning expertise in ad optimization, click-through rate and conversion rate prediction, and click fraud detection. Yuxi earned his degree from the University of Toronto, and published five IEEE transactions and conference papers during his master's research. He finds it enjoyable to crawl data from websites and derive valuable insights. He is also an investment enthusiast.

About the Reviewer

Alberto Boschetti is a data scientist with strong expertise in signal processing and statistics. He holds a PhD in telecommunication engineering and currently lives and works in London. In his work projects, he faces challenges daily, spanning across natural language processing (NLP), machine learning, and distributed processing. He is very passionate about his job and always tries to be updated on the latest developments of data science technologies, attending meetups, conferences, and other events. He is the author of Python Data Science Essentials, Regression Analysis with Python, and Large Scale Machine Learning with Python, all published by Packt.

I would like to thank my family, my friends, and my colleagues. Also, a big thanks to the open source community.

www.PacktPub.com

For support files and downloads related to your book, please visit www.PacktPub.com.

Did you know that Packt offers eBook versions of every book published, with PDF and ePub files available? You can upgrade to the eBook version at www.PacktPub.com and as a print book customer, you are entitled to a discount on the eBook copy. Get in touch with us at service@packtpub.com for more details.

At www.PacktPub.com, you can also read a collection of free technical articles, sign up for a range of free newsletters and receive exclusive discounts and offers on Packt books and eBooks.

https://www.packtpub.com/mapt

Get the most in-demand software skills with Mapt. Mapt gives you full access to all Packt books and video courses, as well as industry-leading tools to help you plan your personal development and advance your career.

Why subscribe?

Fully searchable across every book published by Packt

Copy and paste, print, and bookmark content

On demand and accessible via a web browser

Customer Feedback

Thanks for purchasing this Packt book. At Packt, quality is at the heart of our editorial process. To help us improve, please leave us an honest review on this book's Amazon page at https://www.amazon.com/dp/1783553111.

If you'd like to join our team of regular reviewers, you can e-mail us at customerreviews@packtpub.com. We award our regular reviewers with free eBooks and videos in exchange for their valuable feedback. Help us be relentless in improving our products!

Table of Contents

Credits

Preface

What this book covers

What you need for this book

Who this book is for

Conventions

Reader feedback

Customer support

Downloading the example code

Errata

Piracy

Questions

Getting Started with Python and Machine Learning

What is machine learning and why do we need it?

A very high level overview of machine learning

A brief history of the development of machine learning algorithms

Generalizing with data

Overfitting, underfitting and the bias-variance tradeoff

Avoid overfitting with cross-validation

Avoid overfitting with regularization

Avoid overfitting with feature selection and dimensionality reduction

Preprocessing, exploration, and feature engineering

Missing values

Label encoding

One-hot-encoding

Scaling

Polynomial features

Power transformations

Binning

Combining models

Bagging

Boosting

Stacking

Blending

Voting and averaging

Installing software and setting up

Troubleshooting and asking for help

Summary

Exploring the 20 Newsgroups Dataset with Text Analysis Algorithms

What is NLP?

Touring powerful NLP libraries in Python

The newsgroups data

Getting the data

Thinking about features

Visualization

Data preprocessing

Clustering

Topic modeling

Summary

Spam Email Detection with Naive Bayes

Getting started with classification

Types of classification

Applications of text classification

Exploring naive Bayes

Bayes' theorem by examples

The mechanics of naive Bayes

The naive Bayes implementations

Classifier performance evaluation

Model tuning and cross-validation

Summary

News Topic Classification with Support Vector Machine

Recap and inverse document frequency

Support vector machine

The mechanics of SVM

Scenario 1 - identifying the separating hyperplane

Scenario 2 - determining the optimal hyperplane

Scenario 3 - handling outliers

The implementations of SVM

Scenario 4 - dealing with more than two classes

The kernels of SVM

Choosing between the linear and RBF kernel

News topic classification with support vector machine

More examples - fetal state classification on cardiotocography with SVM

Summary

Click-Through Prediction with Tree-Based Algorithms

Brief overview of advertising click-through prediction

Getting started with two types of data, numerical and categorical

Decision tree classifier

The construction of a decision tree

The metrics to measure a split

The implementations of decision tree

Click-through prediction with decision tree

Random forest - feature bagging of decision tree

Summary

Click-Through Prediction with Logistic Regression

One-hot encoding - converting categorical features to numerical

Logistic regression classifier

Getting started with the logistic function

The mechanics of logistic regression

Training a logistic regression model via gradient descent

Click-through prediction with logistic regression by gradient descent

Training a logistic regression model via stochastic gradient descent

Training a logistic regression model with regularization

Training on large-scale datasets with online learning

Handling multiclass classification

Feature selection via random forest

Summary

Stock Price Prediction with Regression Algorithms

Brief overview of the stock market and stock price

What is regression?

Predicting stock price with regression algorithms

Feature engineering

Data acquisition and feature generation

Linear regression

Decision tree regression

Support vector regression

Regression performance evaluation

Stock price prediction with regression algorithms

Summary

Best Practices

Machine learning workflow

Best practices in the data preparation stage

Best practice 1 - completely understand the project goal

Best practice 2 - collect all fields that are relevant

Best practice 3 - maintain consistency of field values

Best practice 4 - deal with missing data

Best practices in the training sets generation stage

Best practice 5 - determine categorical features with numerical values

Best practice 6 - decide on whether or not to encode categorical features

Best practice 7 - decide on whether or not to select features and if so, how

Best practice 8 - decide on whether or not to reduce dimensionality and if so how

Best practice 9 - decide on whether or not to scale features

Best practice 10 - perform feature engineering with domain expertise

Best practice 11 - perform feature engineering without domain expertise

Best practice 12 - document how each feature is generated

Best practices in the model training, evaluation, and selection stage

Best practice 13 - choose the right algorithm(s) to start with

Naive Bayes

Logistic regression

SVM

Random forest (or decision tree)

Neural networks

Best practice 14 - reduce overfitting

Best practice 15 - diagnose overfitting and underfitting

Best practices in the deployment and monitoring stage

Best practice 16 - save, load, and reuse models

Best practice 17 - monitor model performance

Best practice 18 - update models regularly

Summary

Python Machine Learning By Example

Copyright © 2017 Packt Publishing

All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, without the prior written permission of the publisher, except in the case of brief quotations embedded in critical articles or reviews.

Every effort has been made in the preparation of this book to ensure the accuracy of the information presented. However, the information contained in this book is sold without warranty, either express or implied. Neither the author, nor Packt Publishing, and its dealers and distributors will be held liable for any damages caused or alleged to be caused directly or indirectly by this book.

Packt Publishing has endeavored to provide trademark information about all of the companies and products mentioned in this book by the appropriate use of capitals. However, Packt Publishing cannot guarantee the accuracy of this information.

First published: May 2017

Production reference: 1290517

Published by Packt Publishing Ltd.

Livery Place

35 Livery Street

Birmingham 

B3 2PB, UK.

ISBN 978-1-78355-311-2

www.packtpub.com

Preface

Data science and machine learning are some of the top buzzwords in the technical world today. A resurging interest in machine learning is due to the same factors that have made data mining and Bayesian analysis more popular than ever. This book is your entry point to machine learning.

What this book covers

Chapter 1, Getting Started with Python and Machine Learning, is the starting point for someone who is looking forward to enter the field of ML with Python. You will get familiar with the basics of Python and ML in this chapter and set up the software on your machine.

Chapter 2, Exploring the 20 Newsgroups Dataset with Text Analysis Algorithms, explains important concepts such as getting the data, its features, and pre-processing. It also covers the dimension reduction technique, principal component analysis, and the k-nearest neighbors algorithm.

Chapter 3, Spam Email Detection with Naive Bayes, covers classification, naive Bayes, and its in-depth implementation, classification performance evaluation, model selection and tuning, and cross-validation. Examples such as spam e-mail detection are demonstrated.

Chapter 4, News Topic Classification with Support Vector Machine, covers multiclass classification, Support Vector Machine, and how it is applied in topic classification. Other important concepts, such as kernel machine, overfitting, and regularization, are discussed as well.

Chapter 5, Click-Through Prediction with Tree-Based Algorithms, explains decision trees and random forests in depth over the course of solving an advertising click-through rate problem.

Chapter 6, Click-Through Prediction with Logistic Regression, explains in depth the logistic regression classifier. Also, concepts such as categorical variable encoding, L1 and L2 regularization, feature selection, online learning, and stochastic gradient descent are detailed.

Chapter 7, Stock Price Prediction with Regression Algorithms, analyzes predicting stock market prices using Yahoo/Google Finance data and maybe additional data. Also, it covers the challenges in finance and brief explanations of related concepts.

Chapter 8, Best Practices, aims to foolproof your learning and get you ready for production.

After covering multiple projects in this book, the readers will have gathered a broad picture of the ML ecosystem using Python.

What you need for this book

The following are required for you to utilize this book:

scikit-learn 0.18.0

Numpy 1.1

Matplotlib 1.5.1

NLTK 3.2.2

pandas 0.19.2

GraphViz

Quandl Python API

You can use a 64-bit architecture, 2GHz CPU, and 8GB RAM to perform all the steps in this book. You will require at least 8GB of hard disk space.

Who this book is for

This book is for anyone interested in entering data science with machine learning. Basic familiarity with Python is assumed.

Conventions

In this book, you will find a number of text styles that distinguish between different kinds of information. Here are some examples of these styles and an explanation of their meaning.

Code words in text, database table names, folder names, filenames, file extensions, pathnames, dummy URLs, user input, and Twitter handles are shown as follows: The target_names key gives the newsgroups names.

Any command-line input or output is written as follows:

ls -1 enron1/ham/*.txt | wc -l

3672

ls -1 enron1/spam/*.txt | wc -l

1500

New terms and important words are shown in bold. Words that you see on the screen, for example, in menus or dialog boxes, appear in the text like this: Heterogeneity Activity Recognition Data Set.

Warnings or important notes appear in a box like this.

Tips and tricks appear like this.

Reader feedback

Feedback from our readers is always welcome. Let us know what you think about this book-what you liked or disliked. Reader feedback is important for us as it helps us develop titles that you will really get the most out of.

To send us general feedback, simply e-mail feedback@packtpub.com, and mention the book's title in the subject of your message.

If there is a topic that you have expertise in and you are interested in either writing or contributing to a book, see our author guide at www.packtpub.com/authors.

Customer support

Now that you are the proud owner of a Packt book, we have a number of things to help you to get the most from your purchase.

Downloading the example code

You can download the example code files for this book from your account at http://www.packtpub.com. If you purchased this book elsewhere, you can visit http://www.packtpub.com/support  and register to have the files e-mailed directly to you.

You can download the code files by following these steps:

Log in or register to our website using your e-mail address and password.

Hover the mouse pointer on the SUPPORT tab at the top.

Click on Code Downloads & Errata.

Enter the name of the book in the Search box.

Select the book for which you're looking to download the code files.

Choose from the drop-down menu where you purchased this book from.

Click on Code Download.

Once the file is downloaded, please make sure that you unzip or extract the folder using the latest version of:

WinRAR / 7-Zip for Windows

Zipeg / iZip / UnRarX for Mac

7-Zip / PeaZip for Linux

The code bundle for the book is also hosted on GitHub at https://github.com/PacktPublishing/Python-Machine-Learning-By-Example. We also have other code bundles from our rich catalog of books and videos available at https://github.com/PacktPublishing/. Check them out!

Errata

Although we have taken every care to ensure the accuracy of our content, mistakes do happen. If you find a mistake in one of our books-maybe a mistake in the text or the code-we would be grateful if you could report this to us. By doing so, you can save other readers from frustration and help us improve subsequent versions of this book. If you find any errata, please report them by visiting http://www.packtpub.com/submit-errata, selecting your book, clicking on the Errata Submission Form link, and entering the details of your errata. Once your errata are verified, your submission will be accepted and the errata will be uploaded to our website or added to any list of existing errata under the Errata section of that title.

To view the previously submitted errata, go to https://www.packtpub.com/books/content/support and enter the name of the book in the search field. The required information will appear under the Errata section.

Piracy

Piracy of copyrighted material on the Internet is an ongoing problem across all media. At Packt, we take the protection of our copyright and licenses very seriously. If you come across any illegal copies of our works in any form on the Internet, please provide us with the location address or website name immediately so that we can pursue a remedy.

Please contact us at copyright@packtpub.com with a link to the suspected pirated material.

We appreciate your help in protecting our authors and our ability to bring you valuable content.

Questions

If you have a problem with any aspect of this book, you can contact us at questions@packtpub.com, and we will do our best to address the problem.

Getting Started with Python and Machine Learning

We kick off our Python and machine learning journey with the basic, yet important concepts of machine learning. We will start with what machine learning is about, why we need it, and its evolution over the last few decades. We will then discuss typical machine learning tasks and explore several essential techniques of working with data and working with models. It is a great starting point of the subject and we will learn it in a fun way. Trust me. At the end, we will also set up the software and tools needed in this book.

We will get into details for the topics mentioned:

What is machine learning and why do we need it?

A very high level overview of machine learning

Generalizing with data

Overfitting and the bias variance trade off

Cross validation

Regularization

Dimensions and features

Preprocessing, exploration, and feature engineering

Missing Values

Label encoding

One hot encoding

Scaling

Polynomial features

Power transformations

Binning

Combining models

Bagging

Boosting

Stacking

Blending

Voting and averaging

Installing software and setting up

Troubleshooting and asking for help

What is machine learning and why do we need it?

Machine learning is a term coined around 1960 composed of two words—machine corresponding to a computer, robot, or other device, and learning an activity, or event patterns, which humans are good at.

So why do we need machine learning, why do we want a machine to learn as a human? There are many problems involving huge datasets, or complex calculations for instance, where it makes sense to let computers do all the work. In general, of course, computers and robots don't get tired, don't have to sleep, and may be cheaper. There is also an emerging school of thought called active learning or human-in-the-loop, which advocates combining the efforts of machine learners and humans. The idea is that there are routine boring tasks more suitable for computers, and creative tasks more suitable for humans. According to this philosophy, machines are able to learn, by following rules (or algorithms) designed by humans and to do repetitive and logic tasks desired by a human.

Machine learning does not involve the traditional type of programming that uses business rules. A popular myth says that the majority of the code in the world has to do with simple rules possibly programmed in Cobol, which covers the bulk of all the possible scenarios of client interactions. So why can't we just hire many software programmers and continue programming new rules?

One reason is that defining, maintaining, and updating rules becomes more and more expensive over time. The number of possible patterns for an activity or event could be enormous and therefore exhausting all enumeration is not practically feasible. It gets even more challenging to do so when it comes to events that are dynamic, ever-changing, or evolve in real-time. It