Neural Network Programming with Java

Neural Network Programming with Java
Create and unleash the power of neural networks by implementing professional Java code

 Description
Vast quantities of data are produced every second. In this context, neural networks become a powerful technique to extract useful knowledge from large amounts of raw, seemingly unrelated data. One of the most preferred languages for neural network programming is Java as it is easier to write code using it, and most of the most popular neural network packages around already exist for Java. This makes it a versatile programming language for neural networks.

This book gives you a complete walkthrough of the process of developing basic to advanced practical examples based on neural networks with Java.

You will first learn the basics of neural networks and their process of learning. We then focus on what Perceptrons are and their features. Next, you will implement self-organizing maps using the concepts you've learned. Furthermore, you will learn about some of the applications that are presented in this book such as weather forecasting, disease diagnosis, customer profiling, and characters recognition (OCR). Finally, you will learn methods to optimize and adapt neural networks in real time.
Details:
Chapter 1: Getting Started with Neural Networks
Discovering neural networks
Why artificial neural network?
How neural networks are arranged
The very basic element – artificial neuron
Giving life to neurons – activation function
The fundamental values – weights
An important parameter – bias
The parts forming the whole – layers
Learning about neural network architectures
Monolayer networks
Multilayer networks
Feedforward networks
Feedback networks
From ignorance to knowledge – learning process
Let the implementations begin! Neural networks in practice
Summary
Chapter 2: How Neural Networks Learn
Learning ability in neural networks
How learning helps to solve problems
Learning paradigms
Supervised learning
Unsupervised learning
Systematic structuring – learning algorithm
Two stages of learning – training and testing
The details – learning parameters
Error measurement and cost function 
Examples of learning algorithms
Perceptron
Delta rule
Coding of the neural network learning
Learning parameter implementation
Learning procedure
Class definitions
Two practical examples
Perceptron (warning system)
ADALINE (traffic forecast)
Summary
Chapter 3: Handling Perceptrons
Studying the perceptron neural network
Applications and limitations of perceptrons
Linear separation
Classical XOR case
Popular multilayer perceptrons (MLPs)
MLP properties
MLP weights
Recurrent MLP
MLP structure in an OOP paradigm
Interesting MLP applications
Classification in MLPs
Regression in MLPs
Learning process in MLPs
Simple and very powerful learning algorithm – Backpropagation
Elaborate and potent learning algorithm – Levenberg–Marquardt
Hands-on MLP implementation!
Backpropagation in action
Exploring the code
Levenberg–Marquardt implementation
Practical application – types of university enrolments
Summary
Chapter 4: Self-Organizing Maps
Neural networks' unsupervised way of learning
Some unsupervised learning algorithms
Competitive learning or winner takes all
Kohonen self-organizing maps (SOMs)
One-Dimensional SOM
Two-Dimensional SOM
Step-by-step of SOM learning
How to use SOMs
Coding of the Kohonen algorithm
Exploring the Kohonen class
Kohonen implementation (clustering animals)
Summary
Chapter 5: Forecasting Weather
Neural networks for prediction problems
No data, no neural net – selecting data
Knowing the problem – weather variables
Choosing input and output variables
Removing insignificant behaviors – Data filtering
Adjusting values – data preprocessing
Equalizing data – normalization
Java implementation for weather prediction
Plotting charts
Handling data files
Building a neural network for weather prediction
Empirical design of neural networks
Choosing training and test datasets
Designing experiments
Results and simulations
Summary
Chapter 6: Classifying Disease Diagnosis
What are classification problems, and how can neural networks
be applied to them?
A special type of activation function – Logistic regression
Multiple classes versus binary classes
Comparing the expected versus produced results – the
confusion matrix
Classification measures – sensitivity and specificity
Applying neural networks for classification 
Disease diagnosis with neural networks
Using ANN to diagnose breast cancer
Applying NN for an early diagnosis of diabetes
Summary