Index

A

• abstraction
  • about / Let the implementations begin! Neural networks in practice
• activation function
  • about / Giving life to neurons – activation function
• ADALINE (traffic forecast)
  • about / ADALINE (traffic forecast)
• adaptive neural networks
  • about / Adaptive neural networks
  • implementation / Implementation
• adaptive resonance theory (ART)
  • about / Adaptive neural networks, Adaptive resonance theory
• AND logic / Perceptron (warning system)
• ANN
  • used, for diagnosing breast cancer / Using ANN to diagnose breast cancer
• applied unsupervised learning
  • about / Applied unsupervised learning
  • neural network, of radial basis functions / Neural network of radial basis functions
  • Kohonen neural network / Kohonen neural network
  • types of data / Types of data
• artificial intelligence
  • about / Discovering neural networks
• artificial neural networks (ANNs)
  • about / Discovering neural networks
  • need for / Why artificial neural network?
• artificial neuron
  • about / The very basic element – artificial neuron

B

• backpropagation algorithm
  • about / Simple and very powerful learning algorithm – Backpropagation, Backpropagation in action
• Best Matching Unit (BMU) / Step-by-step of SOM learning
• bias
  • about / An important parameter – bias
• binary classes
  • versus multiple classes / Multiple classes versus binary classes

C

• card credit analysis
  • for customer profiling / Card credit analysis for customer profiling
• categorical data
  • about / Types of data
• classes
  • about / Definition of classes among tons of data
• classification
  • sensitivity measure / Classification measures – sensitivity and specificity
  • specificity measure / Classification measures – sensitivity and specificity
  • neural networks, applying for / Applying neural networks for classification
• classification, in MLPs / Classification in MLPs
• classification problems
  • foundations / What are classification problems, and how can neural networks be applied to them?
• cluster analysis
  • about / Cluster analysis
• cluster evaluation
  • about / Cluster evaluation and validation
• clustering
  • about / Clustering task
• clustering task
  • about / Clustering task
• cluster validation
  • about / Cluster evaluation and validation
• coding, of neural network learning
  • about / Coding of the neural network learning
  • learning parameter implementation / Learning parameter implementation
  • learning procedure / Learning procedure
  • class definitions / Class definitions
• common issues, in neural network implementations / Common issues in neural network implementations
• competitive learning
  • about / Competitive learning or winner takes all
• confusion matrix
  • about / Comparing the expected versus produced results – the confusion matrix
• cost function / Supervised learning, Error measurement and cost function
• customer profiling
  • about / Customer profiling
  • data, preprocessing / Preprocessing data
  • card credit analysis / Card credit analysis for customer profiling

D

• data, types
  • numerical / Types of data
  • categorical / Types of data
• data correlation / Data correlation
• data filtering / Data filtering
• data preprocessing, weather forecasting application
  • about / Adjusting values – data preprocessing
  • data equalizing / Equalizing data – normalization
• data selection, weather forecasting application
  • weather variables / Knowing the problem – weather variables
  • input and output variables, selecting / Choosing input and output variables
  • data filtering / Removing insignificant behaviors – Data filtering
• Davies-Bouldin index
  • about / Cluster evaluation and validation
• defined classes
  • using / Definition of classes among tons of data
• delta rule
  • about / Delta rule
• digit recognition
  • about / Simplifying the task – digit recognition
• digit representation approach
  • about / Approach to digit representation
• dimensionality reduction / Dimensionality reduction
• disease diagnosis, with neural networks
  • about / Disease diagnosis with neural networks
  • ANN, used for diagnosing breast cancer / Using ANN to diagnose breast cancer
  • NN, applying for early diagnosing of diabetes / Applying NN for an early diagnosis of diabetes
• Dunn index
  • about / Cluster evaluation and validation

E

• Eclipse IDE
  • code, running with / Programming and running code with the Eclipse IDE
  • debugging with / Debugging with the Eclipse IDE
• encapsulation
  • about / Let the implementations begin! Neural networks in practice
• epoch / Class definitions
• error measurement / Error measurement and cost function
• Euclidian distance algorithm
  • about / Card credit analysis for customer profiling
• external validation
  • about / External validation, External validation

F

• feedback networks
  • about / Feedback networks
• feedforward networks
  • about / Feedforward networks

H

• hands-on MLP implementation
  • about / Hands-on MLP implementation!
  • backpropagation algorithm / Backpropagation in action
  • code, exploring / Exploring the code

I

• implementation, in Java
  • card credit analysis, for customer profiling / Implementation in Java, Card credit analysis for customer profiling
• inheritance
  • about / Let the implementations begin! Neural networks in practice
• input selection
  • about / Input selection
  • data correlation / Data correlation
  • dimensionality reduction / Dimensionality reduction
  • data filtering / Data filtering

J

• Java implementation, weather forecasting application
  • about / Java implementation for weather prediction
  • charts, plotting / Plotting charts
  • data files, handling / Handling data files
  • neural network, building / Building a neural network for weather prediction
• JFreeChart
  • URL / Plotting charts

K

• Kohonen algorithm
  • coding / Coding of the Kohonen algorithm
• Kohonen class
  • exploring / Exploring the Kohonen class
• Kohonen implementation
  • animals, clustering / Kohonen implementation (clustering animals)
• Kohonen neural network
  • about / Kohonen neural network
• Kohonen self-organizing maps (SOMs)
  • about / Kohonen self-organizing maps (SOMs)
  • 1D SOM / One-Dimensional SOM
  • 2D SOM / Two-Dimensional SOM
  • step-by-step learning / Step-by-step of SOM learning
  • using / How to use SOMs

L

• layers, of neurons
  • about / The parts forming the whole – layers
• learning
  • about / How learning helps to solve problems
  • parameters / The details – learning parameters
• learning, stages
  • training / Two stages of learning – training and testing
  • testing / Two stages of learning – training and testing
• learning ability, in neural networks
  • about / Learning ability in neural networks
• learning algorithms, examples
  • about / Examples of learning algorithms
  • perceptrons / Perceptron
  • delta rule / Delta rule
• learning paradigms
  • about / Learning paradigms
  • supervised learning / Supervised learning
  • unsupervised learning / Unsupervised learning
• learning process, in MLPs
  • about / Learning process in MLPs
  • backpropagation algorithm / Simple and very powerful learning algorithm – Backpropagation
  • Levenberg – Marquardt algorithm / Elaborate and potent learning algorithm – Levenberg–Marquardt
  • Levenberg – Marquardt algorithm / Elaborate and potent learning algorithm – Levenberg–Marquardt
• learning process, neural networks / From ignorance to knowledge – learning process
• Levenberg–Marquardt algorithm
  • about / Elaborate and potent learning algorithm – Levenberg–Marquardt
  • implementation / Levenberg–Marquardt implementation
• logistic regression
  • about / A special type of activation function – Logistic regression

M

• MLP applications
  • about / Interesting MLP applications
  • classification / Interesting MLP applications, Classification in MLPs
  • regression / Interesting MLP applications, Regression in MLPs
• monolayer networks
  • about / Monolayer networks
• multilayer networks
  • about / Multilayer networks
• multilayer perceptrons (MLPs)
  • about / Popular multilayer perceptrons (MLPs)
  • properties / MLP properties
  • weights / MLP weights
  • in OOP paradigm / MLP structure in an OOP paradigm
• multiple classes
  • versus binary classes / Multiple classes versus binary classes

N

• NetBeans
  • installing / Download and install NetBeans
  • download link / Download and install NetBeans
  • environment, setting up / Setting up the NetBeans environment
  • project, importing / Importing a project
  • programming with / Programming and running code with NetBeans
  • code, running with / Programming and running code with NetBeans
  • debugging with / Debugging with NetBeans
• neural network, of radial basis functions
  • about / Neural network of radial basis functions
• neural network architectures
  • about / Learning about neural network architectures
  • monolayer networks / Monolayer networks
  • multilayer networks / Multilayer networks
  • feedforward networks / Feedforward networks
  • feedback networks / Feedback networks
• neural network architectures, applications
  • about / Two practical examples
  • perceptron (warning system) / Perceptron (warning system)
  • ADALINE (traffic forecast) / ADALINE (traffic forecast)
• neural network implementations
  • common issues / Common issues in neural network implementations
• neural networks
  • discovering / Discovering neural networks
  • arranging / How neural networks are arranged
  • learning process / From ignorance to knowledge – learning process
  • implementing / Let the implementations begin! Neural networks in practice
  • learning ability / Learning ability in neural networks
  • empirical design / Empirical design of neural networks
  • about / A special type of activation function – Logistic regression
  • applying, for classification / Applying neural networks for classification
• neural networks (NN)
  • about / Why artificial neural network?
• neural networks, for prediction problems
  • about / Neural networks for prediction problems
• neural networks, in pattern recognition
  • applying / How to apply neural networks in pattern recognition
  • data, preprocessing / Preprocessing the data
• neural networks, of empirical design
  • about / Empirical design of neural networks
  • training and test datasets, selecting / Choosing training and test datasets
  • experiments, designing / Designing experiments
  • results and simulations / Results and simulations
• neural networks unsupervised learning
  • about / Neural networks' unsupervised way of learning
• NN
  • applying, for early diagnosis of diabetes / Applying NN for an early diagnosis of diabetes
• normalization
  • about / Equalizing data – normalization
• numerical data
  • about / Types of data
  • examples / Types of data

O

• objects-oriented programming (OOP)
  • about / Let the implementations begin! Neural networks in practice
• OCR problem
  • about / The OCR problem
  • task, simplifying / Simplifying the task – digit recognition
• online retraining
  • about / Online retraining
  • stochastic online learning / Stochastic online learning
  • implementation / Implementation
  • application / Application
• optical characters, recognizing
  • about / Let the coding begin!
  • data, generating / Generating data
  • neural network, building / Building the neural network
  • trial and error / Testing and redesigning – trial and error
  • results / Results

P

• pattern recognition
  • about / What is pattern recognition all about?
  • defined classes / Definition of classes among tons of data
  • undefined classes / What if the undefined classes are undefined?
• pattern recognition tasks
  • examples / What is pattern recognition all about?
• patterns
  • about / What is pattern recognition all about?
• Pearson coefficient / Data correlation
• perceptron
  • about / Examples of learning algorithms, Perceptron
  • studying / Studying the perceptron neural network
  • applications / Applications and limitations of perceptrons
  • limitations / Applications and limitations of perceptrons
  • linear separation / Linear separation
  • XOR case, analyzing / Classical XOR case
• perceptron (warning system)
  • about / Perceptron (warning system)
• polymorphism
  • about / Let the implementations begin! Neural networks in practice
• practical application
  • types of university enrolments / Practical application – types of university enrolments
• Principal Component Analysis (PCA) / Dimensionality reduction
• Proben1
  • about / Disease diagnosis with neural networks
  • reference link / Disease diagnosis with neural networks
• pseudo algorithm
  • reference link, for source code / Stochastic online learning

R

• Radial basis functions (RBFs) / Neural network of radial basis functions
• recurrent MLP
  • about / Recurrent MLP
• regression, in MLPs / Regression in MLPs

S

• single value decomposition (SVD)
  • about / Types of data
• stochastic online learning
  • about / Stochastic online learning
• structure selection
  • about / Structure selection
• supervised learning
  • about / From ignorance to knowledge – learning process, Supervised learning
• systematic structuring
  • about / Systematic structuring – learning algorithm

U

• undefined classes
  • about / What if the undefined classes are undefined?
• Unified Modeling Language (UML) / Let the implementations begin! Neural networks in practice
• unsupervised learning
  • about / Unsupervised learning
• unsupervised learning algorithms
  • about / Some unsupervised learning algorithms
  • competitive learning / Competitive learning or winner takes all

W

• weather forecasting application
  • data, selecting / No data, no neural net – selecting data
  • data preprocessing / Adjusting values – data preprocessing
  • Java implementation / Java implementation for weather prediction
• weights
  • about / The fundamental values – weights