Transactional Machine Learning with Data Streams and AutoML: Build Frictionless and Elastic Machine Learning Solutions with Apache Kafka in the Cloud Using Python

✍ Scribed by Sebastian Maurice

• Publisher: Apress
• Year: 2021
• Tongue: English
• Leaves: 284
• Edition: 1
• Category: Library

✦ Synopsis

Intermediate-Advanced

✦ Table of Contents

Table of Contents
About the Author
About the Technical Reviewer
Acknowledgments
Introduction
Chapter 1: Introduction: Big Data, Auto Machine Learning, and Data Streams
Structured Data
Semi-structured Data
Unstructured Data
A Quick Take on Big Data
Data Quality
Data Streams
Stream Mining
Auto Machine Learning (AutoML)
Machine Learning Model Building Process
Concluding Remarks
Chapter 2: Transactional Machine Learning
Examining TML
Features of TML
Data Fluidity
Joining Data Streams
Data Stream Standardization
Data Stream Integration with AutoML
Low Code
Data Stream Storage Platform (DSSP)
MAADS-VIPER
Algorithm and Insights Management System (AiMS) Dashboard
AutoML Technology
Unsupervised Learning: Detecting Anomalies
Frictionless Machine Learning
Concluding Remarks
Chapter 3: Overcoming Challenges to ML Adoption
Overview of Challenges
Understanding the Root Causes of Challenges in Adopting Advanced Technologies
Data Decentralization
Lack of Corporate Strategy
Advanced Technology Costs
Choosing ML Use Cases
ML Change Acceptance
Technological Barriers
Skill Gap to Adopting ML
Strategy Gap in Adopting ML
Communication Gap in Adopting ML
Approaches to Addressing the Challenges
Discussion and Path Forward
Chapter 4: The Business Value of Transactional Machine Learning
Conventional Machine Learning (CML)
The TML Opportunity
Core Areas of Value from TML
TML Value Areas (Levers)
Measuring Value from TML Solutions
Choosing the Right TML Use Cases
Benefits and Costs
Risks and Pitfalls
Concluding Remarks
Chapter 5: The Technical Components and Architecture for Transactional Machine Learning Solutions
Overview of a TML Solution
Reference Architecture of a TML Solution
Description of Technical Components
Technical Architecture of a TML Solution
Unsupervised Learning
Communication Process Between Components
Data Flows
Example Architecture
TML Cost Management
Concluding Remarks
Chapter 6: Transactional Machine Learning Solution Template with Streaming Visualization
Overview of TML Solution Template
Template Component Details
Kafka Cloud via Confluent Cloud
VIPER Environment File
VIPER, VIPERviz, and HPDE Setup
Kafka Topics and Data Streams
TML Example Code
Walmart Foot Traffic Prediction and Optimization with TML
Unsupervised Learning for Anomaly Detection
Anomaly Detection on Banking Transactions with TML
Concluding Remarks
Chapter 7: Visualize Your TML Model Insights: Optimization, Predictions, and Anomalies
Streaming Anomaly Detection Visualization
Streaming Prediction Visualization
Streaming Optimization Visualization
AiMS Dashboard
Generic Topics’ Visualization
Visualization with WebSockets
Concluding Remarks
Chapter 8: Evolution and Opportunities for Transactional Machine Learning in Almost Every Industry
Areas of Further Exploration
Faster and More Complex Decision-Making by Machines
Broader Adoption of AutoML Techniques and Processes to Data Streams
Stacking and Chaining Different TML Solutions
Concluding Remarks
Chapter 9: TML Project Planning Approach and Closing Thoughts
TML Technology Stack
TML Project Planning Approach
TML Value Creation
Closing Thoughts
Definitions
References
Index