Mobility Patterns, Big Data and Transport Analytics

Mobility Patterns, Big Data and Transport Analytics provides a guide to the new analytical framework and its relation to big data, focusing on capturing, predicting, visualizing and controlling mobility patterns - a key aspect of transportation modeling. The book features prominent international experts who provide overviews on new analytical frameworks, applications and concepts in mobility analysis and transportation systems. Users will find a detailed, mobility ‘structural’ analysis and a look at the extensive behavioral characteristics of transport, observability requirements and limitations for realistic transportation applications and transportation systems analysis that are related to complex processes and phenomena.

This book bridges the gap between big data, data science, and transportation systems analysis with a study of big data’s impact on mobility and an introduction to the tools necessary to apply new techniques.

The book covers in detail, mobility ‘structural’ analysis (and its dynamics), the extensive behavioral characteristics of transport, observability requirements and limitations for realistic transportation applications, and transportation systems analysis related to complex processes and phenomena. The book bridges the gap between big data, data science, and Transportation Systems Analysis with a study of big data’s impact on mobility, and an introduction to the tools necessary to apply new techniques.

• Guides readers through the paradigm-shifting opportunities and challenges of handling Big Data in transportation modeling and analytics
• Covers current analytical innovations focused on capturing, predicting, visualizing, and controlling mobility patterns, while discussing future trends
• Delivers an introduction to transportation-related information advances, providing a benchmark reference by world-leading experts in the field
• Captures and manages mobility patterns, covering multiple purposes and alternative transport modes, in a multi-disciplinary approach
• Companion website features videos showing the analyses performed, as well as test codes and data-sets, allowing readers to recreate the presented analyses and apply the highlighted techniques to their own data

Transport researchers, practitioners, and consultants, Undergraduate and graduate students in transportation programs, Transport policy makers

1. Big Data and Transport Analytics: An Introduction Constantinos Antoniou, Loukas Dimitriou and Francisco Camara Pereira
   1 Introduction
   2 Book Structure
   References
   Further Reading
   Part I: Methodological
   2. Machine Learning Fundamentals
   Francisco C^amara Pereira and Stanislav S. Borysov
   1 Introduction
   2 A Little Bit of History
   3 Deep Neural Networks and Optimization
   4 Bayesian Models
   5 Basics of Machine Learning Experiments
   6 Concluding Remarks
   References
   Further Reading
   3. Using Semantic Signatures for Social Sensing in Urban Environments
   Krzysztof Janowicz, Grant McKenzie, Yingjie Hu, Rui Zhu and Song Gao
   1 Introduction
   2 Spatial Signatures
   2.1 Spatial Point Pattern
   2.2 Spatial Autocorrelations
   2.3 Spatial Interactions With Other Geographic features
   2.4 Place-Based Statistics
   3 Temporal Signatures
   4 Thematic Signatures
   5 Examples
   5.1 Comparing Place Types
   5.2 Coreference Resolution Across Gazetteers
   5.3 Geoprivacy
   5.4 Temporally Enhanced Geolocation
   5.5 Regional Variation
   5.6 Extraction of Urban Functional Regions
   6 Summary
   References
   4. Geographic Space as a Living Structure for Predicting Human Activities Using Big Data
   Bin Jiang and Zheng Ren
   1 Introduction
   2 Living Structure and the Topological Representation
   3 Data and Data Processing
   4 Prediction of Tweet Locations Through Living Structure
   4.1 Correlations at the Scale of Thiessen Polygons
   4.2 Correlations at the Scale of Natural Cities
   4.3 Degrees of Wholeness or Life or Beauty
   5 Implications on the Topological Representation and Living Structure
   6 Conclusion
   Acknowledgments
   References
   5. Data Preparation
   Kristian Henrickson, Filipe Rodrigues and Francisco Camara Pereira
   1 Introduction
   2 Tools and Techniques
   2.1 Scripting and Statistical Analysis Software
   2.2 Database Management Software
   2.3 Working With Web Data
   3 Probe Vehicle Traffic Data
   3.1 Formats and Protocols
   3.2 Data Characteristics
   3.3 Challenges
   3.4 Data Preparation and Quality Control
   4 Context Data
   4.1 The Role of Context Data
   4.2 Types of Context Data
   4.3 Formats and Data Collection
   4.4 Data Cleaning and Preparation
   References
   6. Data Science and Data Visualization
   Michalis Xyntarakis and Constantinos Antoniou
   1 Introduction
   2 Structured Visualization
   3 Multidimensional Data Visualization Techniques
   3.1 Parallel Coordinates
   3.2 Multidimensional Scaling (MDS)
   3.3 t-Distributed Stochastic Neighbor Embedding for High-Dimensional Data Sets (t-SNE)
   4 Case Studies
   4.1 Experimental Setup
   4.2 Car Characteristics Data Set
   4.3 Congestion on I95
   4.4 Dimensionality Reduction on NYC Taxi Flows
   4.5 Dimensionality Reduction on the NYC Turnstile Data Set
   5 Conclusions
   References
   Further Reading
   7. Model-Based Machine Learning for Transportation
   Inon Peled, Filipe Rodrigues and Francisco Camara Pereira
   1 Introduction
   1.1 Background Concepts
   1.2 Notation
   2 Case Study 1: Taxi Demand in New York City
   2.1 Initial Probabilistic Model: Linear Regression
   2.2 Key Components of MBML
   2.3 Inference
   2.4 Model Improvements
   3 Case Study 2: Travel Mode Choices
   3.1 Improvement: Hierarchical Modeling
   4 Case Study 3: Freeway Occupancy in San Francisco
   4.1 Autoregressive Model
   4.2 State-Space Model
   4.3 Linear Dynamical Systems
   4.4 Common Enhancements to LDS
   4.5 NonLinear Variations on LDS
   5 Case Study 4: Incident Duration Prediction
   5.1 Preprocessing
   5.2 Bag-of-Words Encoding
   5.3 Latent Dirichlet Allocation
   6 Summary
   6.1 Further Reading
   References
   8. Textual Data in Transportation Research: Techniques and Opportunities
   Aseem Kinra, Samaneh Beheshti Kashi, Francisco Camara Pereira, Francois Combes and Werner Rothengatter
   1 Introduction
   2 Big Textual Data, Text Sources, and Text Mining
   2.1 Meaning of Text in the Context of Computational Linguistics
   2.2 Text Mining
   2.3 Text Mining Process Model
   2.4 Textual Data Sources in Transportation
   3 Fundamental Concepts and Techniques in Literature
   3.1 Topic Modeling
   3.2 Word2Vec—Text Embeddings With Deep Learning
   4 Application Examples of Big Textual Data in Transportation
   4.1 Developing Transportation and Logistics Performance Classifiers Using NLTK and Naı¨ve Bayes
   4.2 Understanding the Public Opinion Toward Driverless Cars With Topic Modeling
   4.3 Predicting Taxi Demand in Special Events With Text Embeddings and Deep Learning
   5 Conclusions
   References
   Further Reading
   Part II: Applications
   9. Statewide Comparison of Origin-Destination Matrices Between California Travel Model and Twitter
   Jae Hyun Lee, Adam Davis, Elizabeth McBride and Konstadinos G. Goulias
   1 Introduction
   2 California Statewide Travel Demand Model
   3 Twitter Data
   4 Trip Extraction Methods
   5 Models for Matrix Conversion
   5.1 Tobit Regression Model
   5.2 Latent Class Regression Model
   6 Summary and Conclusion
   References
   10. Transit Data Analytics for Planning, Monitoring, Control, and Information
   Haris N. Koutsopoulos, Zhenliang Ma, Peyman Noursalehi and Yiwen Zhu
   1 Introduction
   2 Measuring System Performance From the Passenger’s Point of View
   2.1 The Individual Reliability Buffer Time (IRBT)
   2.2 Denied Boarding
   3 Decision Support With Predictive Analytics
   3.1 Framework
   3.2 Application: Provision of Crowding Predictive Information
   4 Optimal Design of Transit Demand Management Strategies
   4.1 Framework and Problem Formulation
   4.2 Application: Prepeak Discount Design
   5 Conclusion
   Acknowledgments
   References
   Further Reading
   11. Data-Driven Traffic Simulation Models: Mobility Patterns Using Machine Learning Techniques
   Vasileia Papathanasopoulou, Constantinos Antoniou and Haris N. Koutsopoulos
   1 New Modeling Challenges and Data Opportunities
   1.1 New Modeling Requirements
   1.2 New Data Sources
   1.3 Future Challenges
   2 Background
   3 Data-Driven Traffic Performance Modeling: Overall Framework
   3.1 Modeling Approach
   3.2 Model Components
   4 Application to Mesoscopic Modeling
   4.1 Data and Experimental Design
   4.2 Case Study Setup
   4.3 Application and Results
   5 Application to Microscopic Traffic Modeling
   5.1 Data and Experimental Design
   5.2 Case Study Setup
   5.3 Application and Results
   6 Application to Weak Lane Discipline Modeling
   6.1 Data and Experimental Design
   6.2 Case Study Setup
   6.3 Application and Results
   7 Network-Wide Application
   7.1 Implementation Aspects
   7.2 Case Study Setup
   7.3 Results
   8 Conclusions
   Acknowledgments
   References
   12. Big Data and Road Safety: A Comprehensive Review
   Katerina Stylianou, Loukas Dimitriou and Mohamed Abdel-Aty
   1 Introduction
   2 The Role of Big Data in Traffic Safety Analysis
   2.1 Real-Time Crash Prediction
   2.2 Driving Behavior
   3 ADAS and Autonomous Vehicles (AVs)
   4 Conclusions
   References
   13. A Back-Engineering Approach to Explore Human Mobility Patterns Across Megacities Using Online Traffic Maps
   Vana Gkania and Loukas Dimitriou
   1 Introduction
   2 Data and Traffic Information Extraction Methods
   2.1 Cities Characteristics
   2.2 Data Gathering and Preprocessing
   2.3 Extracting Traffic Information by Image Processing
   3 Temporal and Spatiotemporal Mobility Patterns
   3.1 Temporal Patterns
   3.2 Spatiotemporal Patterns
   4 Dynamic Clustering and Propagation of Congestion
   5 Conclusions
   References
   14. Pavement Patch Defects Detection and Classification Using Smartphones, Vibration Signals and Video Images
   Symeon E. Christodoulou, Charalambos Kyriakou and George Hadjidemetriou
   1 Introduction
   2 Brief Literature Review
   2.1 Vibration-Based Methods
   2.2 Vision-Based Methods
   3 Methodology
   3.1 Anomaly Detection Using ANNs and Timeseries Analysis of Vibration Signals
   3.2 Anomaly Detection Using Entropic-Filter Image Segmentation
   3.3 Patch Detection and Measurement Using Support Vector Machines (SVM)
   4 Conclusions
   References
   15. Collaborative Positioning for Urban Intelligent Transportation Systems (ITS) and Personal Mobility (PM): Challenges and Perspectives
   Vassilis Gikas, Guenther Retscher and Allison Kealy
   1 Introduction
   2 C-ITS in Support of the Smart Cities Concept
   2.1 Scientific and Policy Perspectives of Urban C-ITS
   2.2 Taxonomy of Urban C-ITS Applications
   3 User Requirements for Urban C-ITS
   3.1 Requirements Overview
   3.2 Positioning Requirements and Parameters Definition
   4 Positioning Technologies for Urban ITS
   4.1 Radio Frequency-Based (RF) Technologies
   4.2 MEMS-Based Inertial Navigation
   4.3 Optical Technologies
   5 Measuring Types and Positioning Techniques
   5.1 Absolute Positioning Techniques
   5.2 Relative and Hybrid Positioning Techniques
   6 CP for C-ITS
   6.1 From Single Sensor Positioning to CP
   6.2 Fusion Algorithms and Techniques for CP
   7 Application Cases of Integrated Urban C-ITS
   7.1 Case 1: Smart-Bike Systems as a Component of Urban C-ITS
   7.2 Case 2: Smart Intersection for Traffic Control and Safety
   8 Discussion, Perspectives, and Conclusions
   References
   Further Reading
   Conclusions