Intelligence at the Edge

Intelligence at the Edge

by Michael Harvey
Released February 2020
Publisher(s): SAS Institute
ISBN: 9781642957785

Read it now on the O’Reilly learning platform with a 10-day free trial.

O’Reilly members get unlimited access to books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.

Start your free trial

Book description

Explore powerful SAS analytics and the Internet of Things!

The world that we live in is more connected than ever before. The Internet of Things (IoT) consists of mechanical and electronic devices connected to one another and to software through the internet. Businesses can use the IoT to quickly make intelligent decisions based on massive amounts of data gathered in real time from these connected devices. IoT increases productivity, lowers operating costs, and provides insights into how businesses can serve existing markets and expand into new ones.

Intelligence at the Edge: Using SAS with the Internet of Things is for anyone who wants to learn more about the rapidly changing field of IoT. Current practitioners explain how to apply SAS software and analytics to derive business value from the Internet of Things. The cornerstone of this endeavor is SAS Event Stream Processing, which enables you to process and analyze continuously flowing events in real time. With step-by-step guidance and real-world scenarios, you will learn how to apply analytics to streaming data. Each chapter explores a different aspect of IoT, including the analytics life cycle, monitoring, deployment, geofencing, machine learning, artificial intelligence, condition-based maintenance, computer vision, and edge devices.

Table of contents

  1. Contents
  2. Preface
    1. About the Internet of Things
    2. About This Book
    3. We Want to Hear from You
  3. About the Author
  4. Chapter 1: Using SAS Event Stream Processing to Process Real World Events
    1. Introduction
    2. How Does SAS Event Stream Processing Work?
    3. What is a SAS Event Stream Processing Model?
    4. Processing Events in Derived Windows
    5. Examples of Event Transformations
      1. Example: Using a Join Window
      2. Example: Using a Pattern Window and a Notification Window
    6. Streaming Analytics
      1. Using SAS Micro Analytic Service Modules with Streaming Analytics
    7. Addressing Big Data and the Internet of Things
      1. Edge Model to Process Measurements from a Power Substation
      2. On-Premises Model for Further Processing
    8. Conclusion
    9. About the Contributors
  5. Chapter 2: Linking Real-World Data to SAS Event Stream Processing Through Connectors and Adapters
    1. Introduction
      1. Choosing Between a Connector or an Adapter
      2. The Role of Third-party Libraries
      3. Loading Connectors
      4. Message Formats Used by Connectors and Adapters
      5. Configuring Connectors and Adapters
    2. Publishers and Subscribers
      1. Publisher Source Window Schema Requirements
      2. Building Events in Publishers
      3. Parsing Events in Subscribers
    3. Writing Your Own Connector
    4. Orchestrating Connectors
    5. Alternative Client Transports for Adapters
    6. Connectors and Adapters Available with SAS Event Stream Processing
    7. Example: Using a File and Socket Connector and a WebSocket Connector
    8. Conclusion
    9. About the Contributor
  6. Chapter 3: Applying Analytics to Streaming Data
    1. Introduction
    2. The Multi-Phase Analytics Life Cycle
    3. Online and Offline Models
    4. Online Versus Offline Model Deployment
    5. Potential for Model Application
    6. Stability Monitoring
    7. Support Vector Data Description
    8. Application of Offline Models on Streaming Data
      1. Stability Monitoring Method
      2. Stability Monitoring Results
      3. Support Vector Data Description Method
      4. Support Vector Data Description Results
    9. Subspace Tracking
    10. Conclusion
    11. References
    12. About the Contributors
  7. Chapter 4: Administering SAS Event Stream Processing Environments with SAS Event Stream Manager
    1. Introduction
    2. Monitoring Your SAS Event Stream Processing Environment
    3. Executing Projects from SAS Event Stream Manager
    4. Governing and Testing Assets
    5. Handling Changes to ESP Servers
    6. Integrating with SAS Model Manager
    7. Accommodating Different User Roles
    8. Example: Deploying a Project Using a Job Template
      1. Prepare the Example Files for Use
      2. Create the Stock Trade Deployment
      3. Add an ESP Server
      4. Upload and Publish the Stock Trade Project
      5. Upload the Stock Trade Job Template
      6. Deploy the Stock Trade Job Template
      7. Monitor the Deployment
      8. Stop the Stock Trade Job
    9. Conclusion
    10. About the Contributor
  8. Chapter 5: SAS Event Stream Processing in an IoT Reference Architecture
    1. What is an IoT Reference Architecture?
      1. IoT Challenges
    2. IoT Reference Architecture Components
    3. Deployment Considerations
      1. Edge Technologies
      2. Cloud Technologies
    4. Use Case
      1. Using SAS Visual Data Mining and Machine Learning to Build a Model
      2. Choosing a Champion Model
      3. Deploying the Model to SAS Event Stream Processing
      4. Monitoring Model Performance
      5. Rebuilding Models
    5. Conclusion
    6. References
    7. About the Contributors
  9. Chapter 6: Artificial Intelligence and the Internet of Things
    1. Introduction
    2. What Do We Mean by Artificial Intelligence?
    3. How Does AI Interact with the Internet of Things?
      1. Increasing Numbers of Smart Connected Devices
      2. New Infonomics of Accumulated Smart Device Metadata
      3. Moving from Traditional to Edge to Mesh Computing
      4. Applications: Integrating AI Technologies with IoT
    4. There’s No Place Like Home: AI and IoT
    5. Creating and Remotely Deploying a SAS Deep Learning Image Detection and Classification Model
      1. Initialize Python Libraries and Launch SAS CAS
      2. Load and Explore the Training Data
      3. Prepare the Training Data for Modeling
      4. Specify Predefined Model Architecture and Import Pre-Trained Model Weights
      5. Train the Model
      6. Score the Test Data to Validate Model Accuracy
      7. Browse the Scored Data
      8. Save Model as ASTORE for Deployment
      9. Upload the ASTORE to SAS CAS
      10. Use SAS ESPPy to Deploy ASTORE Model and Score Streaming Data
    6. What Will the Future Bring?
      1. IoT Governance
      2. 5G Networking
    7. Conclusion
    8. References
    9. About the Contributor
    10. Acknowledgment
  10. Chapter 7: Using Geofences with SAS Event Stream Processing
    1. What Is a Geofence?
    2. Understanding the Geofence Window
    3. Geometries
      1. Polygons
      2. Circles
      3. Polylines
    4. Example
      1. Set Up the Environment
      2. Create a Class to Contain GPS Data
      3. Load the SAS Event Stream Processing Project
      4. Create Connections to Collect Data
      5. Create a Map
      6. Create a Display for the Geofence
      7. Define a Publisher
    5. Conclusion
    6. Reference
    7. About the Contributor
    8. Acknowledgments
  11. Chapter 8: Using Deep Learning with Your IoT Digital Twin
    1. Introduction
    2. How Can Analytics Be Used to Create a Digital Twin?
    3. Digital Twin Examples
      1. Smart Grid
      2. Connected Vehicle
      3. Smart Building
      4. Sensors Might Be Too Expensive
      5. Sensors Might Interfere with the Device
      6. Sensors Might Not Communicate Well
      7. Data Might Be Collected at Different Intervals
      8. Analytic Techniques to Fill the Gaps
    4. Anomaly Detection
      1. Using Your System Model for Anomaly Detection
      2. Operating Modes for Anomaly Detection
      3. Changes in Relationships Between the Parts of Your System
      4. Changes in Patterns Over Time
    5. Predicting the Future with Your Digital Twin Model
    6. Using Your Digital Twin Model for Simulations
    7. Building Your Digital Twin Model
    8. Applying Deep Learning Techniques
    9. Real-time Application of Deep Learning in Your Digital Twin
    10. Applying Computer Vision Techniques
      1. Implementing a Computer Vision Model
    11. Applying Recurrent Neural Networks
    12. Applying Reinforcement Learning Techniques
    13. Hyperparameter Tuning
    14. Conclusion
    15. References
    16. About the Contributor
  12. Chapter 9: Leveraging ESP to Adapt to Variable Data Quality for Location-Based Use Cases
    1. Introduction
      1. Why Use Real-time Location Analytics?
      2. Location and Privacy
      3. Types of Location Data
      4. Location Collection Technologies
    2. Use Cases
      1. Asset Recovery
      2. Customer Engagement
      3. Staffing Priorities
      4. Space Utilization
      5. Cause and Effect
      6. Customer Profiles
    3. Data Variability
      1. Potential Problems in Data Collection
    4. Leveraging SAS Event Stream Processing to Adapt
      1. Model to Passively Track Staff and Customers
    5. Conclusion
    6. About the Contributor
  13. Chapter 10: Condition Monitoring Using SAS Event Stream Processing
    1. Introduction
    2. Experimental Setup
    3. Time Domain Analysis of Vibration Data
    4. Monitoring Specific Frequencies Using Digital Filters
      1. Real-time Analysis Using SAS Event Stream Processing
    5. Monitoring the Whole Fourier Spectrum
    6. Monitoring the Whole Fourier Spectrum by Segments
    7. Conclusion
    8. References
    9. About the Contributors
  14. Chapter 11: Analytics with Computer Vision on the Edge
    1. Introduction
    2. Computer Vision with Deep Learning
    3. Advantages of Real-time Analytics on the Edge
    4. Computer Vision Applications in the IoT
      1. Manufacturing
      2. Government
      3. Data Management for Video Surveillance
      4. Transportation
      5. Health Care
      6. Utility
      7. Financial Services
      8. Retail
      9. Data for Good
      10. Safety Compliance
      11. Personal Protective Equipment Verification
    5. Conclusion
    6. References
    7. About the Contributors
  15. Summary
    1. IoT Partner Ecosystems
    2. Additional Resources

Product information

  • Title: Intelligence at the Edge
  • Author(s): Michael Harvey
  • Release date: February 2020
  • Publisher(s): SAS Institute
  • ISBN: 9781642957785