Radial basis neural network optimization using fruit fly

Inhaltsverzeichnis (Table of Contents)

• Abstract
• List of Figures
• List of Tables
• List of Abbreviations
• Preface
• Chapter 1 Introduction
  • Introduction
  • Motivation and Aim
  • Disposition
• Chapter 2 Artificial Neural Network
  • Biological Neuron
  • Artificial Neural Network
    • Single Layer Perceptron
    • Multi Layer Perceptron
  • Artificial Neuron and Activation Function
    • Linear Activation Function
    • Non Linear Activation Functions
  • Radial Basis Net
  • Learning Methods
    • Supervised Learning
    • Unsupervised learning
  • Neural Network Applications
    • Process Control
    • Speech Recognition
    • Image Compression
    • Medical Imaging
    • Image Processing
    • Face Recognition
    • Road and Obstacle Recognition

  Zielsetzung und Themenschwerpunkte (Objectives and Key Themes)

  This thesis explores the optimization of Radial Basis Neural Networks (RBNNs) using the Fruit Fly Optimization Algorithm (FFOA). The main objective is to demonstrate the effectiveness of the FFOA in improving the performance of RBNNs for various applications. Key themes of the work include: * **Radial Basis Neural Networks (RBNNs):** Understanding the structure, learning mechanisms, and applications of RBNNs. * **Fruit Fly Optimization Algorithm (FFOA):** Exploring the principles and implementation of the FFOA for optimization problems. * **Performance Evaluation:** Assessing the effectiveness of FFOA-optimized RBNNs through various benchmark datasets and applications. * **Comparison with Traditional Methods:** Comparing the performance of the proposed FFOA-based RBNN approach with other optimization methods. * **Applications in Real-World Scenarios:** Investigating the applicability of the optimized RBNN model in real-world domains such as image processing, control systems, and pattern recognition.

  Zusammenfassung der Kapitel (Chapter Summaries)

  **Chapter 1: Introduction** This chapter sets the stage for the thesis, introducing the concept of RBNNs and outlining the motivation behind using FFOA for optimization. It also discusses the disposition of the thesis and the organization of subsequent chapters. **Chapter 2: Artificial Neural Network** This chapter provides a comprehensive overview of artificial neural networks, starting with the fundamental concept of biological neurons. It delves into various types of ANNs, including single-layer perceptrons and multi-layer perceptrons. The chapter also explores different activation functions used in ANNs, focusing on both linear and nonlinear functions. Finally, it discusses various learning methods, including supervised and unsupervised learning, and explores the numerous applications of artificial neural networks across different domains.

  Schlüsselwörter (Keywords)

  The primary focus of this thesis is on Radial Basis Neural Networks, Fruit Fly Optimization Algorithm, optimization, machine learning, artificial neural networks, pattern recognition, image processing, and control systems. This work highlights the potential of combining the strengths of RBNNs and the FFOA to achieve enhanced performance in various applications.

  Frequently Asked Questions

  What is an RBF neural network?

  A Radial Basis Function (RBF) neural network is a type of artificial neural network that uses radial basis functions as activation functions, known for fast convergence and simple structure.

  How does the Fruit Fly Optimization Algorithm (FFOA) work?

  FFOA is a metaheuristic optimization method inspired by the food-finding behavior of fruit flies, characterized by its simplicity and ability to avoid local optima.

  Why is variable selection important in neural networks?

  Variable selection helps remove irrelevant data, reduces multicollinearity, and improves the overall precision and stability of the model.

  What are the main applications of optimized RBNNs?

  They are used in process control, speech and face recognition, image processing, and medical imaging.

  What is the "Mean Impact Value" (MIV)?

  MIV is a metric used in this research to evaluate the influence of independent variables on the dependent variable to facilitate effective variable selection.