Applications of Neural NetworksAlan Murray Applications of Neural Networks gives a detailed description of 13 practical applications of neural networks, selected because the tasks performed by the neural networks are real and significant. The contributions are from leading researchers in neural networks and, as a whole, provide a balanced coverage across a range of application areas and algorithms. The book is divided into three sections. Section A is an introduction to neural networks for nonspecialists. Section B looks at examples of applications using `Supervised Training'. Section C presents a number of examples of `Unsupervised Training'. For neural network enthusiasts and interested, open-minded sceptics. The book leads the latter through the fundamentals into a convincing and varied series of neural success stories -- described carefully and honestly without over-claiming. Applications of Neural Networks is essential reading for all researchers and designers who are tasked with using neural networks in real life applications. |
Contents
NEURAL ARCHITECTURES AND ALGORITHMS | 1 |
12 The Single Layer Perceptron SLP | 2 |
13 The MultiLayer Perceptron MLP | 8 |
14 Radial Basis Function Networks | 14 |
15 Kohonens SelfOrganising Feature Map Networks | 17 |
16 Alternative Training Approaches for Layered Networks | 21 |
17 Summary | 28 |
FACE FINDING IN IMAGES | 35 |
ELECTRICAL LOAD FORECASTING | 157 |
2 THE LOAD FORECASTING PROBLEM | 159 |
3 KOHONENS SELFORGANISING FEATURE MAPS | 163 |
4 APPLICATION OF SELFORGANISING FEATURE MAPS TO DAY TYPE IDENTIFICATION | 167 |
5 APPLICATION OF MULTILAYER FEEDFORWARD ANN TO PEAK LOAD AND VALLEY LOAD PREDICTION | 177 |
6 CONCLUSIONS | 181 |
ON THE APPLICATION OF ARTIFICIAL NEURAL NETWORKS TO PROCESS CONTROL | 191 |
2 FEEDFORWARD ARTD7ICIAL NEURAL NETWORKS | 193 |
2 GENERATION OF FEATURE MAPS | 37 |
3 FEATURE LOCATION IN THE HIGH RESOLUTION IMAGE | 53 |
4 EXPERIMENTS WITH FRESHLY GRABBED SEQUENCES | 58 |
Sex Recognition from Faces Using Neural Networks | 71 |
2 METHODS | 75 |
3 RESULTS | 82 |
4 DISCUSSION | 83 |
ANN BASED CLASSIFICATION OF ARRHYTHMIAS | 93 |
2 DATA PREPROCESSING AND FEATURE EXTRACTION | 96 |
3 SINGLE CHAMBER CLASSIFICATION | 98 |
4 DUAL CHAMBER BASED CLASSIFICATION | 106 |
5 MICROELECTRONIC IMPLEMENTATIONS | 107 |
6 CONCLUSIONS | 111 |
CLASSIFICATION OF CELLS IN CERVICAL SMEARS | 113 |
2 APPPLICATION OF ARTIFICIAL NEURAL NETWORKS | 114 |
3 THE PAPNET SYSTEM | 115 |
4 DEVISING A WORKING PROTOCOL FOR PAPNETASSISTED SCREENING | 116 |
5 PAPNETASSISTED SCREENING VERSUS CONVENTIONAL SCREENING | 118 |
6 PRACTICAL IMPLICATIONS OF PAPNETASSISTED SCREENING | 120 |
7 PAPNETASSISTED RESCREENING FOR QUALITY CONTROL | 121 |
8 PAPNET FOR DETECTION OF CANCER CELLS IN FALSENEGATIVE SMEARS | 122 |
MULTIPHASE FLOW MONITORING IN OIL PIPELINES | 133 |
2 GAMMA DENSITOMETRY AND MULTIPHASE FLOWS | 135 |
3 GENERATION OF DATASETS | 139 |
4 THE NEURAL NETWORK APPROACH | 142 |
5 PREDICTION OF PHASE FRACTIONS | 144 |
6 EFFECTS OF NOISE ON INPUTS | 146 |
7 NOVELTY DETECTION AND NETWORK VALIDATION | 150 |
8 DISCUSSION | 153 |
3 DYNAMIC ARTIFICIAL NEURAL NETWORKS | 196 |
4 INFERENTIAL ESTIMATION | 199 |
7 CONCLUDING REMARKS | 216 |
NESTED NETWORKS FOR ROBOT CONTROL | 221 |
1 INTRODUCTION | 222 |
2 THE NESTED NETWORK METHOD | 225 |
3 SIMULATION RESULTS | 232 |
4 CONCLUSION | 235 |
ADAPTIVE EQUALISATION USING NEURAL NETWORKS | 241 |
102 TRANSVERSAL EQUALISER | 245 |
103 BAYESIAN TRANSVERSAL EQUALISER | 249 |
104 DECISION FEEDBACK EQUALISER | 257 |
105 A COMPARISON WITH THE MLVA | 260 |
106 CONCLUSIONS | 262 |
TDGAMMON A SELFTEACHING BACKGAMMON PROGRAM | 267 |
TD LEARNING OF BACKGAMMON STRATEGY | 270 |
TDGAMMON 10 | 276 |
4 CURRENT STATUS OF TDGAMMON | 278 |
5 CONCLUSIONS | 282 |
TEMPORAL DIFFERENCE LEARNING A CHEMICAL PROCESS CONTROL APPLICATION | 287 |
2 THE APPLICATION | 291 |
3 DISCUSSION | 297 |
4 CONCLUSION | 298 |
| 305 | |
01 INTRODUCTION | 306 |
02 EEG DATABASE | 308 |
04 UNSUPERVISED LEARNING | 309 |
05 SUPERVISED LEARNING | 315 |
Other editions - View all
Common terms and phrases
achieved adaptive adaptive equaliser algorithm approach arrhythmia artificial neural network backgammon backpropagation basis functions Bayesian biomass bioreactor centres channel classifier clustering complex database day types decision boundary detected equaliser error estimates example face feature points feature vector filter hidden layer hidden units hourly load pattern ICDs IEEE implementation input pattern input space iterations K-Map Kohonen learning samples linear load and valley load forecasting micro-feature Multi-Layer Perceptron multilayer perceptron neurons nonlinear number of hidden optimal output node output units PAPNET PAPNET-assisted parameters patient pattern vector peak load performance phase fractions pixel position prediction problem procedure Prosopagnosia radial basis function reinforcement learning robot screening search region self-organising feature map SexNet signal smears strategy supervised learning TD-Gammon technique temporal Temporal Difference Learning test set tiles training data training set valley load variables weights Wiener filter