Principles of Soft Computing Using Python Programming

Gypsy Nandi, PhD, is an Assistant Professor in the Department of Computer Applications at Assam Don Bosco University, Guwahati, India, and Coordinator of VanitAgrata, an empowerment cell of Assam Don Bosco University that teaches rural women digital awareness skills. She has authored two books on data science and soft computing and has also published extensively on machine learning and related subjects.

• About the Author xiPreface xiii1 Fundamentals of Soft Computing 11.1 Introduction to Soft Computing 11.2 Soft Computing versus Hard Computing 21.3 Characteristics of Soft Computing 41.4 Components of Soft Computing 71.4.1 Fuzzy Computing 71.4.2 Neural Network 101.4.3 Evolutionary Computing 121.4.4 Machine Learning 191.4.5 Other Techniques of Soft Computing 292 Fuzzy Computing 352.1 Fuzzy Sets 372.1.1 Features of Fuzzy Membership Functions 382.2 Fuzzy Set Operations 412.3 Fuzzy Set Properties 422.4 Binary Fuzzy Relation 452.5 Fuzzy Membership Functions 462.6 Methods of Membership Value Assignments 492.7 Fuzzification vs. Defuzzification 582.8 Fuzzy c-Means 623 Artificial Neural Network 753.1 Fundamentals of Artificial Neural Network (ANN) 763.2 Standard Activation Functions in Neural Networks 813.2.1 Binary Step Activation Function 813.2.2 Linear Activation Function 833.2.3 Sigmoid/Logistic Activation Function 843.2.4 ReLU Activation Function 853.2.5 Tanh Activation Function 873.2.6 Leaky ReLU Activation Function 883.2.7 SoftMax Activation Function 903.3 Basic Learning Rules in ANN 913.3.1 Hebbian Learning Rule 933.3.2 Perceptron Learning Rule 933.3.3 Delta Learning Rule 943.3.4 Correlation Learning Rule 943.3.5 Competitive Learning Rule 953.3.6 Outstar Learning Rule 953.4 McCulloch–Pitts ANN Model 963.5 Feed-Forward Neural Network 983.5.1 Single-Layer Perceptron 993.5.2 Multilayer Perceptron 1033.5.3 Radial Basis Function Network 1073.6 Feedback Neural Network 1113.6.1 Self-Organizing Map (SOM) 1113.6.2 Hopfield Neural Network (HNN) 1164 Deep Learning 1234.1 Introduction to Deep Learning 1234.2 Classification of Deep Learning Techniques 1254.2.1 Convolutional Neural Networks 1254.2.2 Recurrent Neural Network (RNN) 1374.2.3 Generative Adversarial Network (GAN) 1444.2.4 Autoencoders 1495 Probabilistic Reasoning 1595.1 Introduction to Probabilistic Reasoning 1595.1.1 Random Experiment 1605.1.2 Random Variables 1605.1.3 Independence 1615.1.4 Sample Space 1625.1.5 Odds and Risks 1625.1.6 Expected Values 1655.2 Four Perspectives on Probability 1655.2.1 The Classical Approach 1665.2.2 The Empirical Approach 1665.2.3 The Subjective Approach 1675.2.4 The Axiomatic Approach 1675.3 The Principles of Bayesian Inference 1685.4 Belief Network and Markovian Network 1715.4.1 Syntax and Semantics 1725.4.1.1 Belief Network 1725.4.1.2 Markovian Network 1725.4.2 Conditional Independence 1725.4.3 Learning Methods of the Networks 1775.5 Hidden Markov Model 1785.6 Markov Decision Processes 1865.7 Machine Learning and Probabilistic Models 1916 Population-Based Algorithms 1976.1 Introduction to Genetic Algorithms 1976.2 Five Phases of Genetic Algorithms 1986.2.1 Population Initialization 1986.2.2 Fitness Function Calculation (Evaluation) 1996.2.3 Parent Selection 2006.2.4 Crossover 2046.2.5 Mutation 2056.3 How Genetic AlgorithmWorks? 2076.4 Application Areas of Genetic Algorithms 2126.4.1 Using GA in Travelling Salesman Problem 2126.4.2 Using GA in Vehicle Routing Problem 2166.5 Python Code for Implementing a Simple Genetic Algorithm 2216.6 Introduction to Swarm Intelligence 2256.7 Few Important Aspects of Swarm Intelligence 2276.7.1 Collective Sorting 2286.7.2 Foraging Behavior 2286.7.3 Stigmergy 2296.7.4 Division of Labor 2296.7.5 Collective Transport 2306.7.6 Self-Organization 2316.8 Swarm Intelligence Techniques 2336.8.1 Ant Colony Optimization 2336.8.1.1 How ACO TechniqueWorks? 2336.8.1.2 Applying ACO to Optimization Problems 2356.8.1.3 Using ACO in Travelling Salesman Problem 2366.8.1.4 Python Code for Implementing ACO in TSP 2396.8.2 Particle Swarm Optimization 2416.8.2.1 How PSO TechniqueWorks? 2426.8.2.2 Applying PSO to Optimization Problems 2436.8.2.3 Using PSO in Job-Shop Scheduling Problem 2466.8.2.4 Python Code for Implementing PSO 2477 Rough Set Theory 2557.1 The Pawlak Rough Set Model 2557.1.1 Basic Terms in Pawlak Rough Set Model 2567.1.2 Measures of Rough Set Approximations 2607.2 Using Rough Sets for Information System 2627.3 Decision Rules and Decision Tables 2637.3.1 Parameters of Decision Tables 2647.3.1.1 Consistency Factor 2647.3.1.2 Support and Strength 2657.3.1.3 Certainty Factor 2657.3.1.4 Coverage Factor 2667.3.2 Probabilistic Properties of Decision Tables 2667.4 Application Areas of Rough Set Theory 2687.4.1 Classification 2687.4.2 Clustering 2717.4.3 Medical Diagnosis 2737.4.4 Image Processing 2767.4.5 Speech Analysis 2787.5 Using ROSE Tool for RST Operations 2807.5.1 Attribute Discretization 2807.5.2 Finding Lower and Upper Approximations 2818 Hybrid Systems 2898.1 Introduction to Hybrid Systems 2898.2 Neurogenetic Systems 2918.2.1 GA-based Weight Determination of Multilayer Feed-forward Net 2928.2.2 Neuroevolution of Augmenting Topologies (NEAT) 2988.3 Fuzzy-Neural Systems 3028.3.1 Fuzzy Neurons 3038.3.2 Adaptive Neuro-fuzzy Inference System (ANFIS) 3078.4 Fuzzy-Genetic Systems 3118.5 Hybrid Systems in Medical Devices 315Exercises 322Index 327