Learn AI with Python

Explore Machine Learning and Deep Learning techniques for Building Smart AI Systems Using Scikit-Learn, NLTK, NeuroLab, and Keras (English Edition)

Gaurav Leekha

• Publisher: BPB Publications
• Year: 2021
• Topic: Computer Science
• eBook ISBN:
  9789391392611
• Subtopic: Artificial Intelligence (AI) & Semantics
• Index: Computer Science

Build AI applications using Python to intelligently interact with the world around you.

Key Features
? Covers the practical aspects of Machine Learning and Deep Learning concepts with the help of this example-rich guide to Python.
? Includes graphical illustrations of Natural Language Processing and its implementation in NLTK.
? Covers deep learning models such as R-CNN and YOLO for object recognition and teaches how to build an image classifier using CNN.

Description
The book 'Learn AI with Python' is intended to provide you with a thorough understanding of artificial intelligence as well as the tools necessary to create your intelligent applications.This book introduces you to artificial intelligence and walks you through the process of establishing an AI environment on a variety of platforms. It dives into machine learning models and various predictive modeling techniques, including classification, regression, and clustering. Additionally, it provides hands-on experience with logic programming, ASR, neural networks, and natural language processing through real-world examples and fully functional Python implementation. Finally, the book deals with profound models of learning such as R-CNN and YOLO. Object detection in images is also explained in detail using Convolutional Neural Networks (CNNs), which are also explained.By the end of this book, you will have a firm grasp of machine learning and deep learning techniques, as well as a steered methodology for formulating and solving related problems.

What you will learn
? Learn to implement various machine learning and deep learning algorithms to achieve smart results.
? Understand how ML algorithms can be applied to real-life applications.
? Explore logic programming and learn how to use it practically to solve real-life problems.
? Learn to develop different types of artificial neural networks with Python.
? Understand reinforcement learning and how to build an environment and agents using Python.
? Work with NLTK and build an automatic speech recognition system.

Who this book is for
This book is for anyone interested in learning about artificial intelligence and putting it into practice with Python. This book is also valuable for intermediate Machine Learning practitioners as a reference guide. Readers should be familiar with the fundamental understanding of Python programming and machine learning techniques.

Table of Contents
1. Introduction to AI and Python
2. Machine Learning and Its Algorithms
3. Classification and Regression Using Supervised Learning
4. Clustering Using Unsupervised Learning
5. Solving Problems with Logic Programming
6. Natural Language Processing with Python
7. Implementing Speech Recognition with Python
8. Implementing Artificial Neural Network (ANN) with Python
9. Implementing Reinforcement Learning with Python
10. Implementing Deep Learning and Convolutional Neural Network

CHAPTER 1

Introduction to AI and Python

Introduction

What’s the very first thing that comes into your mind when you think of Artificial Intelligence (AI)? It may be an automated machine, robots, or an image of the brain with some processing. If yes, then your understanding of AI is appropriate but vague. So, you may be wondering, what exactly the concept of AI is? This chapter provides a brief overview of AI. It covers various fields of study in AI, real-life applications of AI, and agents and environments. This chapter also addresses the Python programming language, one of the most popular programming languages used by developers today for building AI applications. It also highlights features of Python, its installation, and steps to run the Python script.

Structure

In this chapter, we will discuss the following topics:

• Introduction to Artificial Intelligence (AI)
• Learning AI
• Understanding intelligence
• Various fields of study in AI
• Application of AI in various industries
• How does artificial intelligence learn?
• AI – agents and environments
• AI and Python – how do they relate?
• Python3 – installation and setup

Objectives

After studying this unit, you will understand the basics of AI. You will also learn various fields of study in AI and its applications in various industries. You will be able to install Python 3 on Windows, Linux, and Mac OS X. You will also understand the reason for choosing Python for AI projects.

Introduction to Artificial Intelligence (AI)

John McCarthy, an American computer scientist, who was a pioneer and an inventor, coined the term Artificial Intelligence (AI) in his 1955 proposal for the 1956 Dartmouth Conference, the first artificial intelligence conference. According to him, AI is The science and engineering of making intelligent machines, especially intelligent computer programs.

As we can see, Artificial Intelligence is composed of two words, first is Artificial, which means man-made, and second is Intelligence, which means thinking power. Hence, we can say that AI means a man-made thinking power. We can define AI as:

"A branch of information technology by which we can create intelligent machines that can think like a human, behave like a human, and also able to make the decisions at its own.”

AI is accomplished by studying how humans think, learn, and decide while trying to solve a problem, and then using this outcome as a base for developing intelligent machines. The best part of AI is that we do not have to preprogram a machine, instead we can create a machine with programmed algorithms that can work with its own intelligence.

Why to learn AI?

Are machines capable of thinking? This is a simple question that is very difficult to answer. Different researchers defined terms such as thought or intelligence in different ways. When we look more closely at AI, this is just one of the problems that are encountered.

But, one thing is clear that the current progress in the development of algorithms, combined with greater processing power and exponential growth in the amount of available data, means that AI is now capable of developing systems that can perform tasks that were previously viewed as the exclusive domain of human beings. Some of the capabilities of AI, due to which we should learn it, are as follows:

• AI is capable of learning through data: In our day-to-day life, we deal with huge amounts of data and our mind can’t keep track of such huge data. AI’s capability of learning through data helps us to automate things.
• AI is capable of teaching itself: In this digital era, data itself keep changing at a rapid pace, so the knowledge that is derived from such data must also be updated constantly. To fulfill this purpose, a system should be intelligent, and AI can help us to create such intelligent systems.
• AI can respond in real time: If you use the internet regularly, you’re probably using some real-time applications in the fields of e-commerce, healthcare, retail, manufacturing, self-driving cars, and so on. AI along with the help of neural networks can analyze the data more deeply and hence can respond to the situations that are based on the conditions in real time.
• AI can achieve a greater degree of accuracy: Deep learning, a subset of machine learning, extends the potential of AI to more complex tasks that can only be computed through multiple steps. These tasks are often performed with a greater degree of accuracy.

Understanding intelligence

To build AI applications (smart systems that can think and act like a human), it’s necessary to understand the concept of intelligence. As discussed before, different researchers defined terms such as thought or intelligence in different ways. Let’s define intelligence keeping in mind the scope of AI:

• Ability to take decisions: From a set of many deciding factors, it’s important to take the optimal, correct, and accurate decisions. This measures intelligence in a generic way as well as in terms of AI.
• Ability to prove results: Another important factor that measures intelligence is the ability to prove that why this decision has been chosen.
• Ability to think logically: Do you think, in this world, everything can be proved by mathematical formulae or proof? No, as humans, for many things, we need to apply our common sense, think logically, and conclude. This ability also measures intelligence.
• Ability to learn and improve: How do we develop our experiences? Whenever we learn something new, we develop our experiences. These experiences help all of us to make better decisions and better opportunities in the future. This also measures intelligence in a generic way as well as in terms of AI because the more we learn from the external environment, the more we have the ability to improve ourselves.

Types of intelligence

According to Howard Gardner¹, an American development psychologist, there are eight multiple intelligences.

• Linguistic–verbal intelligence: It is the ability to speak, recognize, and use the mechanism of phonology, syntax, and semantics. Some of the characteristics of people with linguistic–verbal intelligence are:
  • They enjoy reading and writing.
  • They can explain things very well.
  • They are good at debating or giving persuasive speeches.
  • They are also good at remembering written and spoken information.
  For example, writers, narrators, teachers, and journalists.
• Musical intelligence: It is the ability to create, communicate, and understand pitch, rhythm, and meaning of sounds. Some of the characteristics of people with musical intelligence are:
  • They enjoy singing as well as playing musical instruments.
  • They can recognize musical patterns and tones easily.
  • They are good at remembering songs and melodies.
  • They have a great understanding of musical structure, rhythm, and notes.
  For example, musicians, singers, music teachers, and composers.
• Logical–mathematical intelligence: It is the ability to use, understand relationships in the absence of action or objects, and understand complex as well as abstract ideas. Some of the characteristics of people with logical–-mathematical intelligence are:
  • They have excellent problem-solving skills.
  • They enjoy thinking about abstract ideas.
  • They are good at solving scientific experiments.
  • They also like conducting scientific experiments.
  For example, mathematicians, engineers, computer programmers, and scientists.
• Visual–spatial intelligence: It is the ability to perceive visual information, change it, re-create images without reference to the objects, construct 3-dimensional images, move, and rotate them. Some of the characteristics of people with visual–spatial intelligence are:
  • They enjoy drawing and painting.
  • They recognize patterns very easily.
  • They are good at interpreting pictures, graphs, and charts.
  • They are also good at putting puzzles together.
  For example, architects, artists, astronauts, and physicists.
• Bodily kinesthetic intelligence: It is the ability to use part of or complete body to solve problems. It is the control of fine and coarse motor skills. Some of the characteristics of people with bodily kinesthetic intelligence are:
  • They have excellent physical coordination.
  • They enjoy creating things by themselves.
  • They are good at dancing and sports.
  For example, players, builders, actors, and dancers.
• Intra-personal intelligence: It is the ability to distinguish among one’s feelings, intentions, and motivations. Some of the characteristics of people with intra-personal intelligence are:
  • They are good at analyzing their strengths and weaknesses.
  • They enjoy analyzing and learning through theories and ideas.
  • They have excellent self-awareness.
  • They understand the basis for their motivations as well as feelings.
  For example, writers, theorists, and scientists.
• Inter-personal intelligence: Unlike intra-personal intelligence, it is the ability to recognize and make distinctions among other feelings, beliefs, and intentions. Some of the characteristics of people with interpersonal intelligence are:
  • They are good at communicating verbally.
  • They are also skilled at nonverbal communication.
  • They always tend to create a positive relationship with others.
  • They are also good at resolving conflicts in groups.
  For example, psychologists, philosophers, and politicians.
• Naturalistic intelligence: It is the ability to explore the environment and learning about other species. The indiv...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Dedication Page
• About the Author
• About the Reviewer
• Acknowledgement
• Preface
• Errata
• Table of Contents
• 1. Introduction to AI and Python
• 2. Machine Learning and Its Algorithms
• 3. Classification and Regression Using Supervised Learning
• 4. Clustering Using Unsupervised Learning
• 5. Solving Problems with Logic Programming
• 6. Natural Language Processing with Python
• 7. Implementing Speech Recognition with Python
• 8. Implementing Artificial Neural Network (ANN) with Python
• 9. Implementing Reinforcement Learning with Python
• 10. Implementing Deep Learning and Convolutional Neural Network
• Index