Cover image for Graph-Powered Machine Learning

Graph-Powered Machine Learning

Author Alessandro Negro
Release Date: 2021/09/01
ISBN: 9781617295645
Topic:
0%
27
Chapters
0-1
Hours read
0k
Total Words
    Start Reading Now    
       Add to Wishlist       
View table of contents

Graph-Powered Machine Learning teaches you how to exploit the natural relationships in structured and unstructured datasets using graph-oriented machine learning algorithms and tools. In this authoritative book, you’ll master the architectures and design practices of graphs, and avoid common pitfalls. Author Alessandro Negro explores examples from real-world applications that connect GraphML concepts to real world tasks.

Table of Contents

  1. Graph-Powered Machine Learning
  2. Copyright
  3. dedication
  4. contents
  5. front matter
    1. foreword
    2. preface
    3. acknowledgments
    4. about this book
    5. Who should read this book?
    6. How this book is organized
    7. About the code
    8. liveBook discussion forum
    9. Online resources
    10. about the author
    11. about the cover illustration
  6. Part 1 Introduction
  7. 1 Machine learning and graphs: An introduction
    1. 1.1 Machine learning project life cycle
    2. 1.1.1 Business understanding
    3. 1.1.2 Data understanding
    4. 1.1.3 Data preparation
    5. 1.1.4 Modeling
    6. 1.1.5 Evaluation
    7. 1.1.6 Deployment
    8. 1.2 Machine learning challenges
    9. 1.2.1 The source of truth
    10. 1.2.2 Performance
    11. 1.2.3 Storing the model
    12. 1.2.4 Real time
    13. 1.3 Graphs
    14. 1.3.1 What is a graph?
    15. 1.3.2 Graphs as models of networks
    16. 1.4 The role of graphs in machine learning
    17. 1.4.1 Data management
    18. 1.4.2 Data analysis
    19. 1.4.3 Data visualization
    20. 1.5 Book mental model
    21. Summary
    22. References
  8. 2 Graph data engineering
    1. 2.1 Working with big data
    2. 2.1.1 Volume
    3. 2.1.2 Velocity
    4. 2.1.3 Variety
    5. 2.1.4 Veracity
    6. 2.2 Graphs in the big data platform
    7. 2.2.1 Graphs are valuable for big data
    8. 2.2.2 Graphs are valuable for master data management
    9. 2.3 Graph databases
    10. 2.3.1 Graph database management
    11. 2.3.2 Sharding
    12. 2.3.3 Replication
    13. 2.3.4 Native vs. non-native graph databases
    14. 2.3.5 Label property graphs
    15. Summary
    16. References
  9. 3 Graphs in machine learning applications
    1. 3.1 Graphs in the machine learning workflow
    2. 3.2 Managing data sources
    3. 3.2.1 Monitor a subject
    4. 3.2.2 Detect a fraud
    5. 3.2.3 Identify risks in a supply chain
    6. 3.2.4 Recommend items
    7. 3.3 Algorithms
    8. 3.3.1 Identify risks in a supply chain
    9. 3.3.2 Find keywords in a document
    10. 3.3.3 Monitor a subject
    11. 3.4 Storing and accessing machine learning models
    12. 3.4.1 Recommend items
    13. 3.4.2 Monitoring a subject
    14. 3.5 Visualization
    15. 3.6 Leftover: Deep learning and graph neural networks
    16. Summary
    17. References
  10. Part 2 Recommendations
    1. References
  11. 4 Content-based recommendations
    1. 4.1 Representing item features
    2. 4.2 User modeling
    3. 4.3 Providing recommendations
    4. 4.4 Advantages of the graph approach
    5. Summary
    6. References
  12. 5 Collaborative filtering
    1. 5.1 Collaborative filtering recommendations
    2. 5.2 Creating the bipartite graph for the User-Item dataset
    3. 5.3 Computing the nearest neighbor network
    4. 5.4 Providing recommendations
    5. 5.5 Dealing with the cold-start problem
    6. 5.6 Advantages of the graph approach
    7. Summary
    8. References
  13. 6 Session-based recommendations
    1. 6.1 The session-based approach
    2. 6.2 The events chain and the session graph
    3. 6.3 Providing recommendations
    4. 6.3.1 Item-based k-NN
    5. 6.3.2 Session-based k-NN
    6. 6.4 Advantages of the graph approach
    7. Summary
    8. References
  14. 7 Context-aware and hybrid recommendations
    1. 7.1 The context-based approach
    2. 7.1.1 Representing contextual information
    3. 7.1.2 Providing recommendations
    4. 7.1.3 Advantages of the graph approach
    5. 7.2 Hybrid recommendation engines
    6. 7.2.1 Multiple models, single graph
    7. 7.2.2 Providing recommendations
    8. 7.2.3 Advantages of the graph approach
    9. Summary
    10. References
  15. Part 3 Fighting fraud
  16. 8 Basic approaches to graph-powered fraud detection
    1. 8.1 Fraud prevention and detection
    2. 8.2 The role of graphs in fighting fraud
    3. 8.3 Warm-up: Basic approaches
    4. 8.3.1 Finding the origin point of credit card fraud
    5. 8.3.2 Identifying a fraud ring
    6. 8.3.3 Advantages of the graph approach
    7. Summary
    8. References
  17. 9 Proximity-based algorithms
    1. 9.1 Proximity-based algorithms: An introduction
    2. 9.2 Distance-based approach
    3. 9.2.1 Storing transactions as a graph
    4. 9.2.2 Creating the k-nearest neighbors graph
    5. 9.2.3 Identifying fraudulent transactions
    6. 9.2.4 Advantages of the graph approach
    7. Summary
    8. References
  18. 10 Social network analysis against fraud
    1. 10.1 Social network analysis concepts
    2. 10.2 Score-based methods
    3. 10.2.1 Neighborhood metrics
    4. 10.2.2 Centrality metrics
    5. 10.2.3 Collective inference algorithms
    6. 10.3 Cluster-based methods
    7. 10.4 Advantages of graphs
    8. Summary
    9. References
  19. Part 4 Taming text with graphs
  20. 11 Graph-based natural language processing
    1. 11.1 A basic approach: Store and access sequence of words
    2. 11.1.1 Advantages of the graph approach
    3. 11.2 NLP and graphs
    4. 11.2.1 Advantages of the graph approach
    5. Summary
    6. References
  21. 12 Knowledge graphs
    1. 12.1 Knowledge graphs: Introduction
    2. 12.2 Knowledge graph building: Entities
    3. 12.3 Knowledge graph building: Relationships
    4. 12.4 Semantic networks
    5. 12.5 Unsupervised keyword extraction
    6. 12.5.1 Keyword co-occurrence graph
    7. 12.5.2 Clustering keywords and topic identification
    8. 12.6 Advantages of the graph approach
    9. Summary
    10. References
  22. appendix A. Machine learning algorithms taxonomy
    1. A.1 Supervised vs. unsupervised learning
    2. A.2 Batch vs. online learning
    3. A.3 Instance-based vs. model-based learning
    4. A.4 Active vs. passive learning
    5. Reference
  23. appendix B. Neo4j
    1. B.1 Neo4j introduction
    2. B.2 Neo4j installation
    3. B.2.1 Neo4j server installation
    4. B.2.2 Neo4j Desktop installation
    5. B.3 Cypher
    6. B.4 Plugin installation
    7. B.4.1 APOC installation
    8. B.4.2 GDS Library
    9. B.5 Cleaning
    10. References
  24. appendix C. Graphs for processing patterns and workflows
    1. C.1 Pregel
    2. C.2 Graphs for defining complex processing workflows
    3. C.3 Dataflow
    4. References
  25. appendix D. Representing graphs
    1. References
  26. index
3.12.161.77