A Design and Development Method for Artificial Neural Network Projects

Inhaltsangabe: Abstract: In the 1980s research efforts and successes made artificial neural networks popular. Since the 1990s engineers have been using this foundation for problem solving. But artifiial neural network solutions for "real-world" problems are sometimes hard to find because of the complexity of the domain and because of the vast number of design attributes the engineer has to deal with. This thesis provides a structured overview of attributes in the design process of artificial neural networks and reviews technical process models. Current development methods for artificial neural networks are then reviewed and critiqued. The thesis concludes with a new design and development method for artificial neural networks. Inhaltsverzeichnis: Table of Contents: List of figuresx List of tablesxi Introduction1 1.Design attributes in ANN3 1.1ANN models4 1.1.1Node level7 1.1.2Network level9 1.1.3Training level9 1.2Data and data representation10 1.3Global system design12 1.4Hardware and software implementation13 1.5Characteristics of ANNs15 1.5.1Advantages of ANNs15 1.5.2Limitations and concerns16 2.Technical process models and engineering methods18 2.1Why use an engineering method?18 2.2Evolutionary model of engineering discipline20 2.3Overview of technical process models22 2.3.1Taxonomy of technical process models24 2.3.2Prototyping25 2.3.3Incremental method26 2.3.4Strict contractual approach26 2.3.5Deciding on process models and methods26 2.3.6Examples of process models27 2.3.7Representation of process models27 2.4Quality criteria of process models29 3.Current engineering methods for ANNs30 3.1Why a special method for ANNs?30 3.1.1Are conventional engineering methodologies suitable for ANNs?30 3.2Methods for expert systems31 3.3System identication methods35 3.4Bailey and Thompson37 3.4.1Criticism43 3.5Medsker and Liebowitz44 3.6Jones and Franklin45 3.7Schalko47 3.8Karayiannis and Nicolaos48 3.8.1Criticism49 3.9Nelson and