Essential Data Analytics, Data Science, and AI: A Practical Guide for a Data-Driven World

Contents....5 About the Author....13 About the Technical Reviewers....14 Foreword....17 Chapter 1: Introduction....19 Data Analytics....19 Data Science....20 Artificial Intelligence....20 The Dream Team....21 What to Expect?....21 Chapter 2: Obtaining Data....24 Understanding the Different Types of Data....24 Structured Data....24 Unstructured Data....25 Quantitative Data....26 Qualitative Data....26 Time Series Data....27 Cross-Sectional Data....27 Spatial Data....28 Ways to Obtain Data....28 Web Scraping: Extracting Data from the Web....28 API: Application Programming Interface....28 Questionnaires and Surveys....29 Popular Sites and Datasets....29 Kaggle....29 Hugging Face....31 Quandl....34 Synthetic Data....35 Telemetry....38 Case Studies and Real-World Examples....38 Obtaining Real-World Data....38 Retail Sales Forecasting....38 Financial Fraud Detection....39 Supply Chain Optimization....39 Obtaining Synthetic Data....39 Preserving Privacy in Healthcare....39 Image Recognition....40 Ethics and Liabilities....40 CCPA: California Consumer Privacy Act....40 GDPR: General Data Protection Regulation....41 HIPAA: Health Insurance Portability and Accountability Act....41 Enhancing Transparency and Understanding Using Dataset Cards....41 Why Are Data Cards Important?....42 Conclusion....43 Chapter 3: ETL Pipeline....44 Introduction to ETL....45 Importance of ETL....45 ETL Tools and Techniques....49 Case Studies and Real-World Examples....61 Retail Analytics....61 Healthcare Data Integration....61 Social Media....61 Fraud Detection in Financial Transactions....62 Smart Home Automation....62 Conclusion....63 Chapter 4: Exploratory Data Analysis....64 Introduction to Exploratory Data Analysis....64 Importance of Exploratory Data Analysis....66 Exploratory Data Analysis Tools and Techniques....76 Python....76 R....77 Tableau....77 Power BI....77 Choosing the Right Visualizations for EDA....83 Use Cases of EDA for Business....85 Retail Industry....85 Healthcare....86 Finance....88 Manufacturing....89 Conclusion....90 Chapter 5: Machine Learning Models....91 Introduction to Machine Learning Models....93 Deep Learning Models....98 Key Concepts and Terminology....101 Importance of Machine Learning Models....106 Healthcare....106 Finance....106 Retail....106 Transportation....107 Marketing....107 Machine Learning Models’ Tools and Techniques....108 Programming Languages....108 Frameworks and Libraries....108 TensorFlow....108 PyTorch....108 Keras....108 scikit-Learn....109 Transformers....109 Integrated Development Environments....109 Jupyter Notebook....109 Google Colab....110 Feature Selection and Engineering....114 Pre-trained vs. Fine-Tuned Models....115 Prompt Engineering....115 Conclusion....115 Chapter 6: Evaluating Models....117 Introduction to Evaluating Models....117 Importance of Evaluating Models....118 Tools and Techniques When Evaluating Models’....120 Examples with Code....133 Monitoring Your Model....136 Examples Where Machine Learning Models Can Be Useful....137 Conclusion....137 Chapter 7: When to Use Machine Learning Models....139 Introduction to When to Use Machine Learning Models....139 Importance of Understanding When to Use Machine Learning Models....140 Assessing the Necessity and Value of Machine Learning Models....141 Identifying the Right Problems....141 Navigating Open-Ended Challenges....142 Adapting to Change in Dynamic Environments....143 Solving Tough Problems....143 Critical Conditions for Successful Machine Learning Models....144 Critical Conditions for Successful AI Products....144 Enhancing Your Model by Leveraging the Data from Interactions with It....145 Cost Considerations for AI Products....146 When to Consider Alternatives....146 Tools and Techniques of When to Use Machine Learning Models’....147 Techniques and Tools....148 Examples Where Machine Learning Models Can Be Useful....155 Finance: Fraud Detection....155 Retail: Personalized Recommendations....155 Conclusion....155 Chapter 8: Where Machine Learning Models Live....156 Introduction to Where Machine Learning Models Live....156 Key Considerations to Think About When Deciding Where Your Machine Learning Models Should Live....157 Deciding Where the Machine Learning Should Live....162 Built-In Machine Learning Model....162 Cloud-Based Machine Learning Model....163 Stored Machine Learning Results....163 Mixing Different Approaches....164 Tools and Techniques....164 Cloud Services....164 On-Premise Options....165 Edge Devices....165 Hybrid Systems....166 Examples of Machine Learning Deployment Strategies for Industry Needs....166 Retail: Cloud-Based Recommendation Systems....166 Healthcare: Edge AI for Real-Time Diagnostics....167 Manufacturing: Hybrid Systems for Predictive Maintenance....167 Conclusion....168 Chapter 9: Telemetry....169 Introduction to Telemetry....169 Importance of Telemetry....170 Telemetry Tools and Techniques....171 Data Collection Tools....171 Data Transmission Tools....171 Data Analysis Tools....172 Model Decay, a.k.a. Model Drift....172 Types of Drift....173 Ways to Detect Drift....174 Examples of How Telemetry Can Be Used in the World....175 E-commerce Recommendation Systems....175 Autonomous Vehicles....177 Healthcare Monitoring System....178 Conclusion....179 Chapter 10: Adversaries and Abuse....180 Introduction to Adversaries and Abuse....180 Importance of Understanding Adversaries and Abuse in Machine Learning....182 Impact on Security and Privacy....182 Economic and Operational Implications....183 Trust in Machine Learning Systems....184 Ways Adversarial Attacks Can Be Used....184 Data Poisoning in Autonomous Vehicles....184 Model Inversion in Healthcare....184 Evasion Attacks Using Adversarial Images....184 Mitigating Risk....187 Ethical Considerations....188 Conclusion....193 Chapter 11: Working with Models....194 Introduction....194 The Historical Context of Technological Advancement....194 The Benefits of AI: Transformative Potential Across Industries....195 Healthcare: Improved Diagnostics and Personalized Medicine....195 Education: Personalized Learning....196 Manufacturing: Automation and Predictive Maintenance....196 Everyday Life: Personal Assistants and Smart Home Devices....196 AI Enhancing Human Capabilities....197 Successful AI Integration....197 Healthcare....197 Finance....198 Consumer Goods....198 Addressing Common Fears and Misconceptions....198 AI As a Threat to Privacy: Balancing Innovation with Ethical Considerations....198 The Myth of AI As an Uncontrollable Force....199 Responsible AI....199 Understanding Bias....199 Mitigating Bias – Human in the Loop Approach....204 Ways Humans Can Be in the Loop but Not Limited To....205 Benefits of Humans in the Loop....205 Challenges of Humans in the Loop....206 Mitigating Bias – Why Diversity Matters....206 Building Trust and Transparency – Ethical AI: Ensuring Fairness, Accountability, and Transparency....206 The Role of Education and Public Engagement....207 The Future AI: Opportunities and Challenges....208 Conclusion....209