Introducing Machine Learning

Master machine learning concepts and develop real-world solutions Machine learning offers immense opportunities, and Introducing Machine Learning delivers practical knowledge to make the most of them. Dino and Francesco Esposito start with a quick overview of the foundations of artificial intelligence and the basic steps of any machine learning project. Next, they introduce Microsoft’s powerful ML.NET library, including capabilities for data processing, training, and evaluation. They present families of algorithms that can be trained to solve real-life problems, as well as deep learning techniques utilizing neural networks. The authors conclude by introducing valuable runtime services available through the Azure cloud platform and consider the long-term business vision for machine learning. ·        14-time Microsoft MVP Dino Esposito and Francesco Esposito help you ·         Explore what’s known about how humans learn and how intelligent software is built ·         Discover which problems machine learning can address ·         Understand the machine learning pipeline: the steps leading to a deliverable model ·         Use AutoML to automatically select the best pipeline for any problem and dataset ·         Master ML.NET, implement its pipeline, and apply its tasks and algorithms ·         Explore the mathematical foundations of machine learning ·         Make predictions, improve decision-making, and apply probabilistic methods ·         Group data via classification and clustering ·         Learn the fundamentals of deep learning, including neural network design ·         Leverage AI cloud services to build better real-world solutions faster     About This Book ·         For professionals who want to build machine learning applications: both developers who need data science skills and data scientists who need relevant programming skills ·         Includes examples of machine learning coding scenarios built using the ML.NET library

Table of Contents:
 Introduction Part I Laying the Groundwork of Machine Learning Chapter 1 How Humans Learn The Journey Toward Thinking Machines     The Dawn of Mechanical Reasoning     Godel’s Incompleteness Theorems     Formalization of Computing Machines     Toward the Formalization of Human Thought     The Birth of Artificial Intelligence as a Discipline The Biology of Learning     What Is Intelligent Software, Anyway?     How Neurons Work     The Carrot-and-Stick Approach     Adaptability to Changes Artificial Forms of Intelligence     Primordial Intelligence     Expert Systems     Autonomous Systems     Artificial Forms of Sentiment Summary Chapter 2 Intelligent Software Applied Artificial Intelligence     Evolution of Software Intelligence     Expert Systems General Artificial Intelligence     Unsupervised Learning     Supervised Learning Summary Chapter 3 Mapping Problems and Algorithms Fundamental Problems     Classifying Objects     Predicting Results     Grouping Objects More Complex Problems     Image Classification     Object Detection     Text Analytics Automated Machine Learning     Aspects of an AutoML Platform     The AutoML Model Builder in Action Summary Chapter 4 General Steps for a Machine Learning Solution Data Collection     Data-Driven Culture in the Organization     Storage Options Data Preparation     Improving Data Quality     Cleaning Data     Feature Engineering     Finalizing the Training Dataset Model Selection and Training     The Algorithm Cheat Sheet     The Case for Neural Networks     Evaluation of the Model Performance Deployment of the Model     Choosing the Appropriate Hosting Platform     Exposing an API Summary Chapter 5 The Data Factor Data Quality     Data Validity     Data Collection Data Integrity     Completeness     Uniqueness     Timeliness     Accuracy     Consistency What’s a Data Scientist, Anyway?     The Data Scientist at Work     The Data Scientist Tool Chest     Data Scientists and Software Developers Summary Part II Machine Learning In .NET Chapter 6 The .NET Way Why (Not) Python?     Why Is Python So Popular in Machine Learning?     Taxonomy of Python Machine Learning Libraries     End-to-End Solutions on Top of Python Models Introducing ML.NET     Creating and Consuming Models in ML.NET     Elements of the Learning Context Summary Chapter 7 Implementing the ML.NET Pipeline The Data to Start From     Exploring the Dataset     Applying Common Data Transformations     Considerations on the Dataset The Training Step     Picking an Algorithm     Measuring the Actual Value of an Algorithm     Planning the Testing Phase     A Look at the Metrics Price Prediction from Within a Client Application     Getting the Model File     Setting Up the ASP.NET Application     Making a Taxi Fare Prediction     Devising an Adequate User Interface     Questioning Data and Approach to the Problem Summary Chapter 8 ML.NET Tasks and Algorithms The Overall ML.NET Architecture     Involved Types and Interfaces     Data Representation     Supported Catalogs Classification Tasks     Binary Classification     Multiclass Classification Clustering Tasks     Preparing Data for Work     Training the Model     Evaluating the Model Transfer Learning     Steps for Building an Image Classifier     Applying Necessary Data Transformations     Composing and Training the Model     Margin Notes on Transfer Learning Summary Part III Fundamentals of Shallow Learning Chapter 9 Math Foundations of Machine Learning Under the Umbrella of Statistics     The Mean in Statistics     The Mode in Statistics     The Median in Statistics Bias and Variance     The Variance in Statistics     The Bias in Statistics Data Representation     Five-number Summary     Histograms     Scatter Plots     Scatter Plot Matrices     Plotting at the Appropriate Scale Summary Chapter 10 Metrics of Machine Learning Statistics vs. Machine Learning     The Ultimate Goal of Machine Learning     From Statistical Models to Machine Learning Models Evaluation of a Machine Learning Model     From Dataset to Predictions     Measuring the Precision of a Model Preparing Data for Processing     Scaling     Standardization     Normalization Summary Chapter 11 How to Make Simple Predictions: Linear Regression The Problem     Guessing Results Guided by Data     Making Hypotheses About the Relationship The Linear Algorithm     The General Idea     Identifying the Cost Function     The Ordinary Least Square Algorithm     The Gradient Descent Algorithm     How Good Is the Algorithm? Improving the Solution     The Polynomial Route     Regularization Summary Chapter 12 How to Make Complex Predictions and Decisions: Trees The Problem     What’s a Tree, Anyway?     Trees in Machine Learning     A Sample Tree-Based Algorithm Design Principles for Tree-Based Algorithms     Decision Trees versus Expert Systems     Flavors of Tree Algorithms Classification Trees     How the CART Algorithm Works     How the ID3 Algorithm Works Regression Trees     How the Algorithm Works     Tree Pruning Summary Chapter 13 How to Make Better Decisions: Ensemble Methods The Problem The Bagging Technique     Random Forest Algorithms     Steps of the Algorithms     Pros and Cons The Boosting Technique     The Power of Boosting     Gradient Boosting     Pros and Cons Summary Chapter 14 Probabilistic Methods: Naïve Bayes Quick Introduction to Bayesian Statistics     Introducing Bayesian Probability     Some Preliminary Notation     Bayes’ Theorem     A Practical Code Review Example Applying Bayesian Statistics to Classification     Initial Formulation of the Problem     A Simplified (Yet Effective) Formulation     Practical Aspects of Bayesian Classifiers Naïve Bayes Classifiers     The General Algorithm     Multinomial Naïve Bayes     Bernoulli Naïve Bayes     Gaussian Naïve Bayes Naïve Bayes Regression     Foundation of Bayesian Linear Regression     Applications of Bayesian Linear Regression Summary Chapter 15 How to Group Data: Classification and Clustering A Basic Approach to Supervised Classification     The K-Nearest Neighbors Algorithm     Steps of the Algorithm     Business Scenarios Support Vector Machine     Overview of the Algorithm     A Quick Mathematical Refresher     Steps of the Algorithm Unsupervised Clustering     A Business Case: Reducing the Dataset     The K-Means Algorithm     The K-Modes Algorithm     The DBSCAN Algorithm Summary Part IV Fundamentals of Deep Learning Chapter 16 Feed-Forward Neural Networks A Brief History of Neural Networks     The McCulloch-Pitt Neuron     Feed-Forward Networks     More Sophisticated Networks Types of Artificial Neurons     The Perceptron Neuron     The Logistic Neuron Training a Neural Network     The Overall Learning Strategy     The Backpropagation Algorithm Summary Chapter 17 Design of a Neural Network Aspects of a Neural Network     Activation Functions     Hidden Layers     The Output Layer Building a Neural Network     Available Frameworks     Your First Neural Network in Keras     Neural Networks versus Other Algorithms Summary Chapter 18 Other Types of Neural Networks Common Issues of Feed-Forward Neural Networks Recurrent Neural Networks     Anatomy of a Stateful Neural Network     LSTM Neural Networks Convolutional Neural Networks     Image Classification and Recognition     The Convolutional Layer     The Pooling Layer     The Fully Connected Layer Further Neural Network Developments     Generative Adversarial Neural Networks     Auto-Encoders Summary Chapter 19 Sentiment Analysis: An End-to-End Solution Preparing Data for Training     Formalizing the Problem     Getting the Data.     Manipulating the Data     Considerations on the Intermediate Format Training the Model     Choosing the Ecosystem     Building a Dictionary of Words     Choosing the Trainer     Other Aspects of the Network The Client Application     Getting Input for the Model     Getting the Prediction from the Model     Turning the Response into Usable Information Summary Part V Final Thoughts Chapter 20 AI Cloud Services for the Real World Azure Cognitive Services Azure Machine Learning Studio     Azure Machine Learning Service     Data Science Virtual Machines On-Premises Services     SQL Server Machine Learning Services     Machine Learning Server Microsoft Data Processing Services     Azure Data Lake     Azure Databricks     Azure HDInsight     .NET for Apache Spark     Azure Data Share     Azure Data Factory Summary Chapter 21 The Business Perception of AI Perception of AI in the Industry     Realizing the Potential     What Artificial Intelligence Can Do for You     Challenges Around the Corner End-to-End Solutions     Let’s Just Call It Consulting     The Borderline Between Software and Data Science     Agile AI Summary     9780135565667    TOC    12/19/2019