Machine Learning For Beginners

By Mike Jones

This book is a basic introduction to machine learning for absolute beginners .  

 

Will smart machines take over my job? That is a familiar question that comes up whenever people talk about artificial intelligence and machine learning. All of the sudden one's job security is being threatened. People often fear that which they do not know. And this is the goal of this book—to remove that fear of the unknown for the neophytes and complete 100% beginners to artificial intelligence and machine learning. No, you do not need programming background to understand the concepts presented in this book. This is a mere introduction to the vast ocean of knowledge machine learning and artificial intelligence. This book starts with the basic concepts and touches on some complex topics such as: 

• language processing,
• deep learning, and machine language.
• Machine Learning vs. Machine Language
• Machine Learning vs. Natural Language Processing (NLP)
• Machine Learning vs. Artificial Intelligence (AI)
• Machine Learning vs. Deep Learning
• What are Machine Learning Tasks?
• History of Machine Learning
• Types of Machine Learning
• Machine Learning and its Relationship to Other Fields
• An introduction to Computational Learning Theory
• Artificial Neural Networks

You also get a short history lesson on artificial intelligence and later developments. Beginners will also be introduced to algorithms—and that is where most of the fun get turned up a notch. The book goes over a lot of the problems today that machine learning can help solve. And no, the goal is not to replace you in your job. The author breaks everything down for you from linear regression, big data, to artificial neural networks. Will you learn how to program your own machine learning algorithm with this book? Unfortunately that is a whole different beast altogether and will require an entire series of books if you are interested in that subject. What you will learn here from this book may just be the tip of the iceberg but it is enough to dip your toes soaking wet in the subject. Finally you will learn about the tools of the trade and the next steps in case you are interested in the applications of machine learning—either launching a career or incorporating it into your business.

• Language: English
• Publisher: Mike Jones
• Release date: Jun 16, 2021
• ISBN: 9798223405245

Book preview

Introduction

I would like to thank you for downloading this book.

Machine learning is a sub-field of artificial intelligence. You can say that in this field of study we strive to find better ways to apply AI. The goal behind this field of study is to build intelligent machines—those that are able to learn by themselves.

Arthur Samuel described it back in the 90s as the study that gives the ability to the computer for self-learn without being explicitly programmed—that means focusing on dynamic algorithms that analyze data and change as new data comes in. You don’t need to hard code new actions or pathways because the system itself can develop that on its own.

We know that data grows continually and a lot of it is unstructured, which makes the job of analyzing all of that raw data so laborious and at times excruciating. Studies have shown that 80% of today’s data is a cacophony of graphs, documents, photos, videos, and audio. Finding patterns in all of that is a task which has been proven to be impossible for a single human mind.

This is basically one of the major applications of machine learning today—the analysis and computation of massive amounts of data. You can say that because of machine learning technology computers were given a new capability.

And that is the focus of this book. To help you understand what machine learning is in a nutshell. It is assumed that you, the reader, already have some background on related technologies such as computer programming, logic, algorithms, and such. However, you don’t need to be so technical to understand the concepts discussed here.

In fact, we have simplified the terms and concepts that even one who doesn’t have any programming background can the core concepts of machine learning. We’ll go over the details step by step as gently as possible as it were.

Other than going over the different facets of machine learning, you will also learn how to contrast and compare it to different fields of study as well. We’ll touch on natural language processing, artificial intelligence, deep learning, and other related studies and how machine learning is similar to and different from them.

The book will also cover the different algorithms used in machine learning according to its different types. We’ll cover algorithms for supervised learning, unsupervised learning, and reinforcement learning. In other words we’ll go over how machine learning is task driven (e.g. predicting the next value), data driven (e.g. identify and classify customer clusters), and is able to learn from its own mistakes.

We’ll also get a bit technical—just slightly when we cover computational learning theory, big data, statistics, learning and optimization, Bayesian networks, support vector machines, genetic algorithms, and data mining. Again, we have tried to the best of our abilities to simplify these concepts for the lay man.

At the end of this book we have also recommended related AI technologies, open source tools, and programming languages. Well, that is if you are interested to learn how to actually develop this technology or to at least be able to understand its more technical features.

Needless to say, machine learning is a new and exciting field with a lot of beneficial applications. It facilitates more accurate medical diagnosis, it can simplify product marketing, create more accurate sales forecasts, improves the precision of many financial rules, simplifies documentation that is time intensive, fine tune predictive maintenance, and a host of other benefits.

May you develop your own insight into the benefits of machine learning in your own field of study. Again, thank you for downloading this book.

Chapter 1. Just What Is Machine Learning (ML)?

Machine learning is everywhere these days. So many people, which may include you and me, are using it dozens of times every day and yet, and some are not even aware of it.

Machine learning has given us effective web search, practical speech recognition, and self-driving cars. It has even improved our understanding of the human genome on a vast scale that scientists are now at the forefront of studying how medicines affect each individual. This means that someday, medicines will be population-specific, or at its best, patient-specific, which is in contrast to today’s approach: one-size-fits-all.

The resurgence of interest in machine learning has little to do with making human lives convenient, to say the least. It’s slowly becoming popular because many enjoy the ubiquitous home assistants (Amazon Alexa, Google Home, etc.) and superhuman game plays (AlphaGo and Atari with Deep Learning).

Machine learning is increasingly being researched and used due to factors such as affordable data storage, cheaper but more powerful computational processing, and growing varieties and volumes of available data. All of these are popular thanks to Bayesian and data mining analysis. It all has something to do with big data.

All of these factors indicate that it’s now possible to create quickly, if not automatically, software, products, devices, and other technology models that can analyze and deliver bigger, more