Practical AI for Business Leaders, Product Managers, and Entrepreneurs: The Big Data Implementation Handbook

Length: 275 pages
Edition: 1
Language: English
Publisher: de Gruyter
Publication Date: 2022-04-04
ISBN-10: 1501514644
ISBN-13: 9781501514647
Sales Rank: #1096060 (See Top 100 Books)

Implementing advanced analytics at scale in a competitive landscape requires the speed, agility, and mindset of an entrepreneur. It is here that organizations in the future will gain an edge. This book provides the practical guidance　 and theoretical background necessary to implement advanced analytics in mid- to large organizations. The theoretical component draws from the best academic research in management science, computer science, and behavioral economics. The practical component provides design patterns and case studies for implementing big data at scale in complex organizations. The design patterns take the mindset of an entrepreneur having to execute with speed and agility against the backdrop of business, technical, and stakeholder complexity.

Readers will learn:

Principal characteristics of Advanced Analytics and Big Data in the Enterprise Business and technical strategies for fast, affordable, and efficient implementation Strategy and implementation tactics from the point of an “entrepreneur” within a large organization Academic research rationale behind the strategies Best practices in the form of case studies

Title Page
Copyright
Contents
Acknowledgments
Preface
    Why Is AI Important?
    How Is This Book Organized?
    What Do I Need to Know to Be Able to Understand This Book?
    What Is Not in This Book?
    How Should I Read This Book?
    Where Can I Access the Code Examples?
1 Introduction
    1.1 Artificial Intelligence
    1.2 What Is Machine Learning?
    1.3 Areas of Machine Learning
    1.4 Machine Learning Workflow
Part I Machine Learning I
    2 Simple Linear Regression – Concept
        2.1 Bird’s Eye View
        2.2 Fundamental Equation of SLR
        2.3 Some Assumptions of Simple Linear Regression
        2.4 Case Study: Is Earning Related to Learning?
        2.5 Interpreting Regression
        2.6 Summary
    3 Simple Linear Regression – Theory
        3.1 Some Basics of Measuring Distance
        3.2 Total Absolute Error (TAE)
        3.3 Residual Sum of Squares (RSS)
        3.4 Mean Squared Error
        3.5 Analytical vs. Numerical Solutions in Machine Learning
        3.6 Summary
    4 Simple Linear Regression – Practice
        Python libraries
        Import Python libraries
        Load and verify data
        Run Regression – statsmodels
        Review results and model performance
        Make predictions
    5 K-Nearest Neighbors (KNN) – Concept
        5.1 Bird’s Eye View
        5.2 Prediction Structure
        5.3 Case Study: To Loan or Not to Loan
        5.4 Summary
    6 K-Nearest Neighbors (KNN) – Theory
        6.1 Distance Metrics
        6.2 Choosing k in KNN
        6.3 ML Pipelines, Hyperparameters
        6.4 Summary
    7 K-Nearest Neighbors (KNN) – Practice
        Import Python libraries
        Load and verify data
        Run KNN classifier
        Evaluate classifier
        Make predictions
Part II Model Assessment
    8 Model Assessment – Bias-Variance Tradeoff
        8.1 Train-Test Split
        8.2 K-Fold Cross-Validation
        8.3 Underfit and Overfit
        8.4 Bias-Variance Tradeoff
        8.5 Summary
    9 Model Assessment – Regression
        9.1 Mean Squared Error
        9.2 R2 – Intuition
        9.3 R2 – Computation
        9.4 Interpreting R2
        9.5 Summary
    10 Model Assessment – Classification
        10.1 Accuracy
        10.2 Confusion Matrix
        10.3 Precision and Recall
        10.4 ROC/AUC Curve
        10.5 Summary
Part III Machine Learning II
    11 Multiple Linear Regression – Concept
        11.1 Bird’s Eye View
        11.2 Multiple Regression Workflow
        11.3 Case Study: Does Sleep Improve Academic Performance?
        11.4 Interpreting the Regression Equation
        11.5 Summary
    12 Multiple Linear Regression – Theory
        12.1 Standardized Coefficients
        12.2 Multiple Linear Regression Diagnostics Checklist
        12.3 Summary
    13 Multiple Linear Regression – Practice
        Import Python Libraries
        Load and verify data
        Run Regression – statsmodels
        Review results and model performance
        Make predictions
    14 Logistic Regression – Concept
        14.1 Bird’s Eye View
        14.2 Probabilities and Decision Making
        14.3 Case Study: Credit Card Payments
        14.4 Summary
    15 Logistic Regression – Theory
        15.1 Logistic Regression Function
        15.2 Odds, Odds Ratio, and Logit
        15.3 Logistic Regression Equation in Logit Form
        15.4 Interpreting the Coefficients
        15.5 Derivation of the Logistic Regression Equation as Log Odds
        15.6 Summary
    16 Logistic Regression – Practice
        Import Python libraries
        Load and verify data
        Clean data
        Run Regression – statsmodels
        Review results and model performance
        Make predictions
    17 K-Means – Concept
        17.1 Bird’s Eye View
        17.2 Case Study: Marketing Segmentation
        17.3 Summary
    18 K-Means – Theory
        18.1 Expectation – Maximization
        18.2 Choice of K
        18.3 Summary
    19 K-Means – Practice
        Import Python libraries
        Load and verify data
        Run K-Means
        Prepare dataframe for displaying results
        Display results
Part IV Deep Learning
    20 Deep Learning – Bird’s Eye View
        20.1 Foundational Principles
        20.2 From Evidence to Judgment
        20.3 Summary
    21 Neurons
        21.1 Neurons as Functions
        21.2 Mathematics of Neural Computation
        21.3 Linear Algebra for Neural Computation
        21.4 Neural Computation in Matrix Form
        21.5 Summary
    22 Neurons – Practice
        Import Python libraries
        Neuron defined as a set of functions
        Neuron class
    23 Network Architecture
        23.1 Layers
        23.2 Parameters
        23.3 Hyperparameters
        23.4 Summary
    24 Network Architecture – Practice
        Import Keras library
        Define neural network
        Review model
    25 Forward Propagation
        25.1 Forward Propagation Flow
        25.2 Function Composition
        25.3 Forward Propagation Computation
        25.4 Computation Worked Example
        25.5 Summary
    26 Forward Propagation – Practice
        Import libraries
        Define activation functions
        Define class Layer
        Create input values and parameters for Layer1
        Instantiate first and second layers
        Update parameters for Layer 1
        Calculate output for Layers 1 and 2
        Multiple inputs and three layers
    27 Loss Function
        27.1 A Game of Arrows
        27.2 Common Loss Functions in Deep Learning
        27.3 Summary
    28 Loss Function – Practice
        Import libraries
        MSE Loss Function
        Categorical Cross-Entropy Loss Function
        Binary Cross-Entropy Function
    29 Backward Propagation
        29.1 Computation as Gears
        29.2 Gradient Descent
        29.3 Gradient Descent for Simple Linear Regression
        29.4 Gradients and Backpropagation
        29.5 Summary
    30 Backward Propagation – Practice
        Import libraries
        Generate sample data
        Perform regression with scikit-learn
        Perform regression using gradient descent
        Instantiate neuron
    31 Deep Learning – Practice
        Load libraries
        Load and prepare dataset
        Define network
        Compile model and fit data
        Generate predictions and evaluate model
About the Authors
Subject Index