Artificial Intelligence in Highway Safety

Description

Artificial Intelligence in Highway Safety provides cutting-edge advances in highway safety using AI. The author is a highway safety expert. He pursues highway safety within its contexts, while drawing attention to the predictive powers of AI techniques in solving complex problems for safety improvement. This book provides both theoretical and practical aspects of highway safety. Each chapter contains theory and its contexts in plain language with several real-life examples. It is suitable for anyone interested in highway safety and AI and it provides an illuminating and accessible introduction to this fast-growing research trend. Material supplementing the book can be found at https://github.com/subasish/AI_in_HighwaySafety. It offers a variety of supplemental materials, including data sets and R codes. Cover Title Page Copyright Page Dedication Preface Table of Contents List of Abbreviations 1. Introduction 1.1. Highway Safety 1.2. Artificial Intelligence 1.2.1. Idea of Artificial Intelligence 1.2.2. History of AI 1.2.3. Statistical Model vs. AI Algorithm: Two Cultures 1.3. Application of Artificial Intelligence in Highway Safety 1.4. Book Organization 2. Highway Safety Basics 2.1. Introduction 2.2. Influential Factors in Highway Safety 2.3. 4E Approach 2.3.1. Engineering 2.3.2. Education 2.3.3. Enforcement 2.3.4. Emergency 2.4. Intervention Tools 2.5. Data Sources 2.6. Crash Frequency Models 2.7. Crash Severity Models 2.8. Effectiveness of Countermeasures 2.8.1. Observational B/A Studies 2.9. Benefit Cost Analysis 2.10. Transportation Safety Planning 2.11. Workforce Development and Core Competencies 2.11.1. Occupational Descriptors 2.11.2. Core Competencies 3. Artificial Intelligence Basics 3.1. Introduction 3.2. Machine Learning 3.2.1. Supervised Learning 3.2.2. Unsupervised Learning 3.2.3. Semi-supervised Learning 3.2.4. Reinforcement Learning 3.2.5. Deep Learning 3.3. Regression and Classification 3.3.1. Regression 3.3.2. Classification 3.4. Sampling 3.4.1. Probability Sampling 3.4.2. Non-probability Sampling 3.4.3. Population Parameters and Sampling Statistics 3.4.4. Sample Size 4. Matrix Algebra and Probability 4.1. Introduction 4.2. Matrix Algebra 4.2.1. Matrix Multiplication 4.2.2. Linear Dependence and Rank of a Matrix 4.2.3. Matrix Inversion (Division) 4.2.4. Eigenvalues and Eigenvectors 4.2.5. Useful Matrices and Properties of Matrices 4.2.6. Matrix Algebra and Random Variables 4.3. Probability 4.3.1. Probability, Conditional Probability, and Statistical Independence 4.3.2. Estimating Parameters in Statistical Models 4.3.3. Useful Probability Distributions 4.3.4. Mean, Variance and Covariance 5. Supervised Learning 5.1. Introduction 5.2. Popular Models and Algorithms 5.2.1. Logistic Regression 5.2.2. Decision Tree 5.2.3. Support Vector Machine 5.2.4. Random Forests (RF) 5.2.5. Naïve Bayes Classifier 5.2.6. Artificial Neural Networks 5.2.7. Cubist 5.2.8. Extreme Gradient Boosting (XGBoost) 5.2.9. Categorical Boosting (CatBoost) 5.3. Supervised Learning based Highway Safety Studies 6. Unsupervised Learning 6.1. Introduction 6.2. Popular Algorithms 6.2.1. K-Means 6.2.2. K-Nearest Neighbors 6.3. Dimension Reduction Methods in Highway Safety 6.4. Categorical Data Analysis 6.4.1. The Singular Value Decomposition 6.5. Correspondence Analysis 6.5.1. Multiple Correspondence Analysis 6.5.2. Taxicab Correspondence Analysis 6.6. Unsupervised Learning, Semi-Supervised, and Reinforcement Learning based Highway safety Studies 7. Deep Learning 7.1. Introduction 7.2. Popular Algorithms 7.2.1. LSTM 7.2.2. Monte Carlo Sampling 7.3. Boltzmann Machines 7.3.1. Boltzmann Machine Learning 7.3.2. Generative Adversarial Networks 7.4. Deep Learning Categories 7.4.1. Convolutional Neural Networks (CNNs) 7.4.2. CNN Structure 7.4.3. CNN Architectures and Applications 7.4.4. Forward and Backward Propagation 7.4.5. Pretrained Unsupervised Networks 7.4.6. Autoencoders 7.4.7. Deep Belief Network 7.4.8. Recurrent and Recursive Neural Networks 7.5. Deep Learning based Highway Safety Studies 8. Natural Language Processing 8.1. Introduction 8.2. Text Mining 8.3. Topic Modeling 8.3.1. Latent Dirichlet Allocation 8.3.2. Structural Topic Model (STM) 8.3.3. Keyword Assisted Topic Model 8.3.4. Text Summarization 8.4. Sentence Centrality and Centroid-based Summarization 8.5. Centrality-based Sentence Salience 8.5.1. Eigenvector Centrality and LexRank 8.5.2. Continuous LexRank 8.6. NLP Based Highway Safety Studies 9. Explainable AI 9.1. Introduction 9.1.1. Partial Dependence Plot (PDP) 9.1.2. Individual Conditional Expectation (ICE) 9.1.3. Accumulated Local Effects (ALE) Plot 9.1.4. Local Surrogate (LIME) 9.1.5. Shapley Value 9.1.6. SHAP (SHapley Additive exPlanations) 10. Disruptive and Emerging Technologies in Highway Safety 10.1. Introduction 10.2. Risks Associated with Emerging and Disruptive Technologies 10.2.1. Connected and Autonomous Vehicles 10.2.2. Electric Vehicles 10.2.3. Mobility as a Service/Mobility on Demand 10.2.4. Advanced Air Mobility 10.3. Studies on Emerging and Disruptive Technologies 11. Conclusions and Future Needs 11.1. Introduction 11.2. Highway Safety AI 101 11.3. Ethics in Highway Safety AI 11.3.1. Ethics and Regulation 11.3.2. Bias, Fairness, Interpretability, Robustness, and Security 11.3.3. Governance 11.4. AI based Highway Safety Guidances Appendix A: Case Study of Exploratory Data Analysis Appendix B: Steps of Big Data Analysis in Highway Safety Appendix C: ML Interpretability and Model Selection Appendix D: Develop an Interactive Map Appendix E: Develop an interactive Shiny App for Highway Safety Analysis with AI Models Appendix F: Develop an Interactive Shiny App with Application Programming Interface (API) based Queries Appendix G: Alternative to Crash Tree Tool Appendix H: Example of Quick Bibliographic Search Appendix I: Example of Self-Organizing Maps Appendix J: Example of Correspondence Analysis Appendix K: Example of Deep Explainer Appendix L: Road Safety Professional (RSP) Certification Needs Index