Cognitive Computing with IBM Watson: Build smart applications using artificial intelligence as a service

✍ Scribed by Rob High, Tanmay Bakshi

• Publisher: Packt Publishing
• Year: 2019
• Tongue: English
• Leaves: 247
• Category: Library

✦ Synopsis

Understand, design, and create cognitive applications using Watson's suite of APIs.

Key Features

• Develop your skills and work with IBM Watson APIs to build efficient and powerful cognitive apps
• Learn how to build smart apps to carry out different sets of activities using real-world use cases
• Get well versed with the best practices of IBM Watson and implement them in your daily work

Book Description

Cognitive computing is rapidly infusing every aspect of our lives riding on three important fields: data science, machine learning (ML), and artificial intelligence (AI). It allows computing systems to learn and keep on improving as the amount of data in the system grows.

This book introduces readers to a whole new paradigm of computing – a paradigm that is totally different from the conventional computing of the Information Age. You will learn the concepts of ML, deep learning (DL), neural networks, and AI through the set of APIs provided by IBM Watson. This book will help you build your own applications to understand, plan, and solve problems, and analyze them as per your needs. You will learn about various domains of cognitive computing, such as NLP, voice processing, computer vision, emotion analytics, and conversational systems, using different IBM Watson APIs.

From this, the reader will learn what ML is, and what goes on in the background to make computers "do their magic," as well as where these concepts have been applied. Having achieved this, the readers will then be able to embark on their journey of learning, researching, and applying the concept in their respective fields.

What you will learn

• Get well versed with the APIs provided by IBM Watson on IBM Cloud
• Learn ML, AI, cognitive computing, and neural network principles
• Implement smart applications in fields such as healthcare, entertainment, security, and more
• Understand unstructured content using cognitive metadata with the help of Natural Language Understanding
• Use Watson's APIs to create real-life applications to realize their capabilities
• Delve into various domains of cognitive computing, such as media analytics, embedded deep learning, computer vision, and more

Who this book is for

This book is for beginners and novices; having some knowledge about artificial intelligence and deep learning is an advantage, but not a prerequisite to benefit from this book. We explain the concept of deep learning and artificial intelligence through the set of tools IBM Watson provides.

Table of Contents

1. Background, Transition and the Future of Computing
2. Can Machines Converse like Humans?
3. Computer Vision
4. This Is How Computers Speak
5. Expecting Empathy from Dumb Computers
6. Language - How Watson deals with NL
7. Structuring Unstructured Content Through Watson
8. Putting It All Together with Watson
9. Future: Cognitive Computing and You

✦ Table of Contents

Title Page
Copyright and Credits
About Packt
Contributors
Table of Contents
Preface
Background, Transition, and the Future of Computing
Transitioning from conventional to cognitive computing
Limitations of conventional computing
Solving conventional computing's problems
Workings of machine learning
Machine learning and its uses 
Cons of machine learning 
Introduction to IBM Watson
Hardware and software requirements
Signing up for IBM Cloud
Summary 
Can Machines Converse Like Humans?
Creating a conversational agent workspace
Creating an instance of Watson Assistant and a workspace
The sample application
Creating a set of conversational intents
Recognizing entities
Identifying entities through annotators
Building a dialog
Creating the dialog for a complex Intent using Frame Slots
Context variables
Programming your conversation application
Emerging features
Summary
Further reading
Computer Vision
Can machines visually perceive the world around them?
The past – classical computer vision
The present – deep learning for computer vision
Creating a basic image-recognition system
Creating an instance of Watson Visual Recognition and a classifier
Uploading data and training the classifier
Testing the classifier
Creating a Python application to classify with Watson
Handling the case where you don't have training data
Using the facial detection model
Summary
This Is How Computers Speak
A computer that talks
Playing sound through the speaker
Getting fancier with how to speak
Controlling pronunciation
Customizing speech synthesis
Using sounds-like customization
Streaming and timing
A fun application of the speech service
Talking to the computer
Getting voice from a microphone
Using the WebSockets interface to speech recognition
Telephones are not good microphones
More about base models
Dealing with speaker hesitations
Customizing the speech recognition service
Customizing Watson's language model
Customizing the acoustic model for Watson
Leveraging batch processing
Summary
Further reading
Expecting Empathy from Dumb Computers
Introducing empathy
Understanding the complexities of sentiment
The functionality of the Tone Analyzer API
How you can use the Tone Analyzer API
Understanding personality through natural language
Using natural language to infer personality traits
Calling the Personality Insights API
Summary
Language - How Watson Deals with NL
Natural language translation – the past 
Natural language – it's intrinsically unstructured
Natural language translation – the present
Translating between languages with Language Translator
Training custom NMT models with Watson
Categorizing text using Natural Language Classifier
Summary
Further reading
Structuring Unstructured Content Through Watson
Using computers that recognize what you mean
Introducing the NLU service
Alternative sources of literature
Types of analyses
Categories
Concepts
Emotion
Sentiment
Entities
Relations
Keywords
Semantic roles
Parts of speech (syntax)
Customizing NLU
Preparing to annotate
Creating a type system
Adding documents
As an aside
Preparing documents for use in Watson Knowledge Studio
Loading documents into Watson Studio
Performing annotations
Editing the type system
The importance of being thorough
Coreferences
Training Watson
Deploying the custom model to NLU
Using a custom model in NLU
Summary
Putting It All Together with Watson
Recapping Watson Services
Building a sample application from Watson Services
The use case and application
The program flow
Translating voice input
Determining intent
Prompting the user for their input
Setting the document of interest
Summarizing entities and concepts
Identifying an entity of interest
Assessing the personality of the entity
Assessing the tone of the entity
Translating text
Classifying text
Running the program
Setup
Summary
Future - Cognitive Computing and You
Other services and features of Watson
Compare and Comply 
Discovery
Watson Studio
Machine learning
Knowledge catalog
Watson OpenScale
The future of Watson
Advances in AI
Generative adversarial networks
Conversational systems
Deep learning
Edge computing
Bias and ethics in AI
Robotics and embodiment
Quantum computing and AI
The future of AI
Summary 
Another Book You May Enjoy
Index