| |
The book presents a breakdown of each variant of machine learning, how it works and how it is used within certain industries. Also covered are various algorithm types (supervised, unsupervised and so on) during training phases of machine learning. The reader learns that with the right tools any developer or technology professional can glean information from their existing data. The book outlines the key types of machine learning, providing coded solutions for real world examples. There is a strong focus on data preparation and data cleaning, the core fundamental of machine learning. Each chapter includes how the code works and running examples. Coverage includes:
Languages for Machine Learning: Hadoop, Mahout, Weka
Planning for Machine Learning: Data Storage/Data Cleaning
Decision Trees: Types/Working Examples
Bayesian Networks: Types/Working Examples
Artificial Neural Networks: Types/Examples/Working Code
Association Rule Learning
Support Vector Machines: Coded Examples
Clustering
Machine Learning as Batch: Hadoop, Mahout, MapReduce: Examples
Learning in Real Time: RabbitMQ
Introduction Chapter 1 What Is Machine Learning?
History of Machine Learning
Alan Turing
Arthur Samuel
Tom M. Mitchell
Summary Definition
Algorithm Types for Machine Learning
Supervised Learning
Unsupervised Learning
The Human Touch
Uses for Machine Learning
Software
Stock Trading
Robotics
Medicine and Healthcare
Advertising
Retail and E-Commerce
Gaming Analytics
The Internet of Things
Languages for Machine Learning
Python
R
Matlab
Scala
Clojure
Ruby
Software Used in This Book
Checking the Java Version
Weka Toolkit
Mahout
SpringXD Hadoop
Using an IDE
Data Repositories
UC Irvine Machine Learning Repository
Infochimps
Kaggle
Summary
Chapter 2 Planning for Machine Learning
The Machine Learning Cycle
It All Starts with a Question
I Don`t Have Data!
Starting Local
Competitions
One Solution Fits All?
Defining the Process
Planning
Developing
Testing
Reporting
Refining
Production
Building a Data Team
Mathematics and Statistics
Programming
Graphic Design
Domain Knowledge
Data Processing
Using Your Computer
A Cluster of Machines
Cloud-Based Services
Data Storage
Physical Discs
Cloud-Based Storage
Data Privacy
Cultural Norms
Generational Expectations
The Anonymity of User Data
Don`t Cross "The Creepy Line"
Data Quality and Cleaning
Presence Checks
Type Checks
Length Checks
Range Checks
Format Checks
The Britney Dilemma
What`s in a Country Name?
Dates and Times
Final Thoughts on Data Cleaning
Thinking about Input Data
Raw Text
Comma Separated Variables
JSON
YAML
XML
Spreadsheets
Databases
Thinking about Output Data
Don`t Be Afraid to Experiment
Summary
Chapter 3 Working with Decision Trees
The Basics of Decision Trees
Uses for Decision Trees
Advantages of Decision Trees
Limitations of Decision Trees
Different Algorithm Types
How Decision Trees Work
Decision Trees in Weka
The Requirement
Training Data
Using Weka to Create a Decision Tree
Creating Java Code from the Classification Testing the Classifier Code
Thinking about Future Iterations
Summary
Chapter 4 Bayesian Networks
Pilots to Paperclips
A Little Graph Theory
A Little Probability Theory
Coin Flips
Conditional Probability
Winning the Lottery
Bayes Theorem
How Bayesian Networks Work
Assigning Probabilities
Calculating Results
Node Counts
Using Domain Experts
A Bayesian Network Walkthrough
Java APIs for Bayesian Networks
Planning the Network
Coding Up the Network
Summary
Chapter 5 Artificial Neural Networks
What Is a Neural Network?
Artificial Neural Network Uses
High-Frequency Trading
Credit Applications
Data Center Management
Robotics
Medical Monitoring
Breaking Down the Artificial Neural Network
Perceptrons
Activation Functions
Multilayer Perceptrons
Back Propagation
Data Preparation for Artificial Neural Networks
Artificial Neural Networks with Weka
Generating a Dataset
Loading the Data into Weka
Configuring the Multilayer Perceptron
Training the Network
Altering the Network
Increasing the Test Data Size
Implementing a Neural Network in Java
Create the Project
The Code
Converting from CSV to Arff
Running the Neural Network
Summary
Chapter 6 Association Rules Learning
Where Is Association Rules Learning Used?
Web Usage Mining
Beer and Diapers
How Association Rules Learning Works
Support
Confidence
Lift
Conviction
Defining the Process
Algorithms
Apriori
FP-Growth
Mining the Baskets-A Walkthrough
Downloading the Raw Data
Setting Up the Project in Eclipse
Setting Up the Items Data File
Setting Up the Data
Running Mahout
Inspecting the Results
Putting It All Together
Further Development
Summary
Chapter 7 Support Vector Machines
What Is a Support Vector Machine?
Where Are Support Vector Machines Used?
The Basic Classification Principles
Binary and Multiclass Classification
Linear Classifiers
Confidence
Maximizing and Minimizing to Find the Line
How Support Vector Machines Approach Classification
Using Linear Classification
Using Non-Linear Classification
Using Support Vector Machines in Weka
Installing LibSVM
A Classification Walkthrough
Implementing LibSVM with Java
Summary
Chapter 8 Clustering
What Is Clustering?
Where Is Clustering Used?
The Internet
Business and Retail
Law Enforcement
Computing
Clustering Models
How the K-Means Works
Calculating the Number of Clusters in a Dataset
K-Means Clustering with Weka
Preparing the Data
The Workbench Method
The Command-Line Method
The Coded Method
Summary
Chapter 9 Machine Learning in Real Time with Spring XD
Capturing the Firehose of Data
Considerations of Using Data in Real Time
Potential Uses for a Real-Time System
Using Spring XD
Spring XD Streams
Input Sources, Sinks, and Processors
Learning from Twitter Data
The Development Plan
Configuring the Twitter API Developer Application
Configuring Spring XD
Starting the Spring XD Server
Creating Sample Data
The Spring XD Shell
Streams 101
Spring XD and Twitter
Setting the Twitter Credentials
Creating Your First Twitter Stream
Where to Go from Here
Introducing Processors
How Processors Work within a Stream
Creating Your Own Processor
Real-Time Sentiment Analysis
How the Basic Analysis Works
Creating a Sentiment Processor
Spring XD Taps
Summary
Chapter 10 Machine Learning as a Batch Process
Is It Big Data?
Considerations for Batch Processing Data
Volume and Frequency
How Much Data?
Which Process Method?
Practical Examples of Batch Processes
Hadoop
Sqoop
Pig
Mahout
Cloud-Based Elastic Map Reduce
A Note about the Walkthroughs
Using the Hadoop Framework
The Hadoop Architecture
Setting Up a Single-Node Cluster
How Map Reduce Works
Mining the Hashtags
Hadoop Support in Spring XD
Objectives for This Walkthrough
What`s a Hashtag?
Creating the Map Reduce Classes
Performing ETL on Existing Data
Product Recommendation with Mahout
Mining Sales Data
Welcome to My Coffee Shop!
Going Small Scale
Writing the Core Methods
Using Hadoop and Map Reduce
Using Pig to Mine Sales Data
Scheduling Batch Jobs
Summary
Chapter 11 Apache Spark
Spark: A Hadoop Replacement?
Java, Scala, or Python?
Scala Crash Course
Installing Scala
Packages
Data Types
Classes
Calling Functions
Operators
Control Structures
Downloading and Installing Spark
A Quick Intro to Spark
Starting the Shell
Data Sources
Testing Spark
Spark Monitor
Comparing Hadoop MapReduce to Spark
Writing Standalone Programs with Spark
Spark Programs in Scala
Installing SBT
Spark Programs in Java
Spark Program Summary
Spark SQL
Basic Concepts
Using SparkSQL with RDDs
Spark Streaming
Basic Concepts
Creating Your First Stream with Scala
Creating Your First Stream with Java
MLib: The Machine Learning Library
Dependencies
Decision Trees
Clustering
Summary
Chapter 12 Machine Learning with R
Installing R
Mac OSX
Windows
Linux
Your First Run
Installing R-Studio
The R Basics
Variables and Vectors
Matrices
Lists
Data Frames
Installing Packages
Loading in Data
Plotting Data
Simple Statistics
Simple Linear Regression
Creating the Data
The Initial Graph
Regression with the Linear Model
Making a Prediction
Basic Sentiment Analysis
Functions to Load in Word Lists
Writing a Function to Score Sentiment
Testing the Function
Apriori Association Rules
Installing the A Rules Package
The Training Data
Importing the Transaction Data
Running the Apriori Algorithm
Inspecting the Results
Accessing R from Java
Installing the rJava Package
Your First Java Code in R
Calling R from Java Programs
Setting Up an Eclipse Project
Creating the Java/R Class
Running the Example
Extending Your R Implementations
R and Hadoop
The RHadoop Project
A Sample Map Reduce Job in RHadoop
Connecting to Social Media with R
Summary
Appendix A SpringXD Quick Start
Installing Manually
Starting SpringXD
Creating a Stream
Adding a Twitter Application Key
Appendix B Hadoop 1.x Quick Start
Downloading and Installing Hadoop
Formatting the HDFS Filesystem
Starting and Stopping Hadoop
Process List of a Basic Job
Appendix C Useful Unix Commands
Using Sample Data
Showing the Contents: cat, more, and less
Example Command
Expected Output
Filtering Content: grep
Example Command for Finding Text
Example Output
Sorting Data: sort
Example Command for Basic Sorting
Example Output
Finding Unique Occurrences: unique
Showing the Top of a File: head
Counting Words: wc
Locating Anything: find
Combining Commands and Redirecting Output
Picking a Text Editor
Colon Frenzy: Vi and Vim
Nano
Emacs
Appendix D Further Reading
Machine Learning
Statistics
Big Data and Data Science
Hadoop
Visualization
Making Decisions
Datasets
Blogs
Useful Websites
The Tools of the Trade
Index ISBN - 9788126553372
|
| |
Pages : 304
|