Artificial Intelligence for Cognitive Modeling: Theory and Practice

Cover
Half Title
Series Page
Title Page
Copyright Page
Table of Contents
Preface
Acknowledgment
Authors
Part A: Artificial Intelligence and Cognitive Computing: Theory and Concept
Chapter 1: Introduction to Artificial Intelligence
1.1 Introduction
1.2 Intelligent Control
1.3 Expert Systems
1.4 Soft Computing Techniques
1.4.1 Fuzzy Systems
1.4.1.1 Architecture of Fuzzy Logic Systems
1.4.2 Neural Networks
1.4.2.1 Basic Architecture of Neural Network
1.4.3 Genetic Algorithms
1.4.4 Adaptive Neuro-Fuzzy Inference System
1.4.5 Real-Time Systems
References
Chapter 2: Practical Approach of Fuzzy Logic Controller
2.1 Introduction
2.2 Classical Set Properties and Operation
2.2.1 Classical Set
2.3 Properties of Crisp Sets
2.4 Concept of Fuzziness
2.4.1 Fuzzy Set
2.4.2 Operation of Fuzzy Sets [ 21, 22 ]
2.4.2.1 Union
2.4.2.2 Intersection
2.4.2.3 Complement
2.4.2.4 Difference
2.4.3 Properties of Fuzzy Sets
2.4.4 Comparison between Crisp Set or Classical Set and Fuzzy Set
2.4.5 Composition of Fuzzy Set
2.4.6 Properties of Fuzzy Composition
2.4.7 Classical Tolerance Relation
2.4.8 Features of Membership Function
2.4.8.1 Fuzzy Set
2.4.8.2 Features of Fuzzy Sets
2.4.8.3 Classification of Fuzzy Sets
2.5 Fuzzification
2.5.1 Institution
2.5.2 Inference
2.5.3 Rank Ordering
2.5.4 Angular Fuzzy Sets
2.5.5 Neural Network
2.5.5.1 A Training the Neural Network
2.5.5.2 Testing the Neural Network
2.5.6 Genetic Algorithm
2.5.7 Inductive Reasoning
2.6 Defuzzification
2.6.1 Lambda Cut for Fuzzy Sets/Alpha Cut
2.6.2 Max Membership Principle
2.6.3 Centroid Method
2.6.4 Weighted Average Method
2.6.5 Mean–Max Membership or Middle of Maxima
2.6.6 Center of Sum Methods
2.6.7 Center of Largest Area
2.7 Examples for Different Defuzzification Methods
2.7.1 Max Membership Method
2.7.2 Centroid Method
2.7.3 Weighted Average Method
2.7.4 Mean Max Membership
2.7.5 Center of Sums
2.7.6 Center of Largest Area
References
Chapter 3: A Practical Approach to Neural Network Models
3.1 Introduction
3.1.1 Network Topology
3.1.1.1 Feed Forward Network
3.1.1.2 Feedback Network
3.1.2 Adjustments of Weights or Learning
3.1.2.1 Supervised Learning
3.1.2.2 Unsupervised Learning
3.1.2.3 Reinforcement Learning
3.1.3 Activation Functions
3.1.3.1 Type of Activation Function
3.1.4 Learning Rules in Neural Network
3.1.4.1 Hebbian Learning Rule
3.1.4.2 Perceptron Learning Rule
3.1.4.3 Delta Learning Rule
3.1.4.4 Competitive Learning Rule (Winner-takes-all)
3.1.4.5 Outstar Learning Rule
3.1.5 Mcculloch Pitts Neuron
3.1.6 Simple Neural Nets for Pattern Classification
3.1.7 Linear Reparability
3.1.8 Perceptron
3.2 Adaptive Linear Neuron (Adaline)
3.2.1 Multiple Adaptive Linear Neurons (Madaline)
3.2.2 Associative Memory Network
3.2.3 Auto Associative Memory
3.2.4 Hetero Associative Memory
3.2.4.1 Architecture
3.3 Bidirectional Associative Memory
3.4 Self-Organizing Maps: Kohonen Maps
3.5 Learning Vector Quantization (LVQ)
3.6 Counter Propagation Network (CPN)
3.6.1 Full Counter Propagation Network (FCPN)
3.6.2 Forward Only Counter Propagation Network
3.7 Adaptive Resonance Theory (ART)
3.8 Standard Back-Propagation Architecture
3.9 Boltzmann Machine Learning
References
Chapter 4: Introduction to Genetic Algorithm
4.1 Introduction
4.2 Optimization Problems
4.2.1 Steps for Solving the Optimization Problem
4.2.2 Point-to-Point Algorithms (P2P)
4.2.3 A∗ Search Algorithm
4.2.4 Simulated Annealing
4.2.5 Genetic Algorithm (GA)
4.2.5.1 Motivation of GA
4.2.5.2 Basic Terminology
4.2.5.2.1 Crossover
4.2.5.2.2 Fitness
4.2.5.2.3 Mutation
4.2.5.2.4 Selection
4.2.5.2.5 Termination
4.2.5.3 Experiments
4.2.5.4 Parameter Tuning Technique in Genetic Algorithm
4.2.5.5 Strategy Parameters
4.3 Constrained Optimization
4.4 Multimodal Optimization
4.5 Multiobjective Optimization
4.6 Combinatorial Optimization
4.6.1 Differential Evolution
4.6.1.1 Suitability of DE in the Field of Optimization
References
Chapter 5: Modeling of ANFIS (Adaptive Fuzzy Inference System) System
5.1 Introduction
5.2 Hybrid Systems
5.2.1 Sequential Hybrid Systems
5.2.2 Auxiliary Hybrid Systems
5.2.3 Embedded Hybrid Systems
5.3 Neuro-Fuzzy Hybrids
5.3.1 Adaptive Neuro-Fuzzy Interference System (ANFIS)
5.3.1.1 Fuzzy Inference System (FIS)
5.3.1.2 Adaptive Network
5.4 ANFIS Architecture
5.4.1 Hybrid Learning Algorithm
5.4.2 Derivation of Fuzzy Model
5.4.2.1 Extracting the Initial Fuzzy Model
5.4.2.2 Subtractive Clustering Technique
5.4.2.3 Grid Partitioning Technique
5.4.2.4 C-Mean Clustering
References
Chapter 6: Machine Learning Techniques for Cognitive Modeling
6.1 Introduction
6.2 Classification of Machine Learning
6.2.1 Supervised Learning
6.2.1.1 Inductive Learning
6.2.1.2 Learning by Version Space
6.2.1.3 Learning by Decision Tree (DT)
6.2.1.4 Analogical Learning
6.2.2 Unsupervised Learning
6.2.3 Reinforcement Learning
6.2.3.1 Learning Automata
6.2.3.2 Adaptive Dynamic Programming
6.2.3.3 Q-learning
6.2.3.3.1 Basic Q-Learning
6.2.3.3.2 Deep Q-Learning
6.2.3.3.3 Hierarchical Q-Learning
6.2.3.3.4 Double Q-Learning
6.2.3.3.5 Multi-Agent
6.2.3.4 Temporal Difference Learning
6.2.4 Learning by Inductive Logic Programming (ILP)
6.3 Summary
References
Part B: Artificial Intelligence and Cognitive Computing: Practices
Chapter 7: Parametric Optimization of n-Channel JFET Using Bio Inspired Optimization Techniques
7.1 Introduction
7.2 Mathematical Description
7.2.1 Current Equation for JFET
7.2.2 Flower Pollination Algorithm
7.2.3 Objective Function
7.3 Methodology
7.4 Result and Discussion
7.5 Conclusion
References
Chapter 8: AI-Based Model of Clinical and Epidemiological Factors for COVID-19
8.1 Introduction
8.2 Related Work
8.3 Artificial Neural Network Based Model
8.3.1 Modeling of Artificial Neural Network
8.3.1.1 Collection, Preprocessing, and Division of Data
8.3.1.2 Implementation of Neural Network
8.3.2 Performance of Training, Testing, and Validation of Network
8.3.3 Performance Evaluation of Training Functions
8.4 Results and Discussion
8.5 Conclusions
References
Chapter 9: Fuzzy Logic Based Parametric Optimization Technique of Electro Chemical Discharge Micro-Machining ( μ -CDM) Process during Micro-Channel Cutting on Silica Glass
9.1 Introduction
9.2 Development of the Set Up
9.3 Experimental Methodology and Result Analysis
9.3.1 Effects of Process Parameters on MRR, OC, and MD
9.3.2 Determination of Optimized Condition
9.4 Conclusions
References
Chapter 10: Study of ANFIS Model to Forecast the Average Localization Error (ALE) with Applications to Wireless Sensor Networks (WSN)
10.1 Introduction
10.2 System Model
10.2.1 Distance Calculation for Generalization of Optimization Problem
10.2.2 Simulation Setup
10.2.3 Experimental Results and Performance Analysis
10.2.3.1 The Effect of Anchor Density
10.2.3.2 The Effect of Communication Range
10.3 Adaptive Neuro-Fuzzy Inference Architecture
10.3.1 Hybrid Learning ANFIS
10.3.2 ANFIS Training Process
10.4 Result Analysis
10.4.1 Grid Partition Method
10.4.2 Subclustering Method
10.5 Conclusions
References
Chapter 11: Performance Estimation of Photovoltaic Cell Using Hybrid Genetic Algorithm and Particle Swarm Optimization
11.1 Introduction
11.2 Mathematics Model and Objective Function of the Solar Cell
11.2.1 Single Diode Model (SDM)
11.2.2 Double Diode Model (DDM)
11.2.3 PV Module Model
11.3 Objective Function
11.4 Proposed Methodology
11.4.1 Improved Cuckoo Search Optimization
11.5 Results and Discussion
11.5.1 Test Information
11.5.1.1 Fitness Test
11.5.1.2 Reliability Test
11.5.1.3 Computational Efficiency Test
11.5.1.4 Convergence Test
11.5.1.5 Accuracy Test
11.5.2 Overall Efficiency
11.5.3 Validation Between Manufacturer’s Datasheet and Experimental Datasheets
11.5.3.1 Case Study 1: Single Diode Model
11.5.3.2 Case Study 2: Double Diode Model
11.6 Conclusions
Conflict of Interest
References
Chapter 12: Bio Inspired Optimization Based PID Controller Tuning for a Non-Linear Cylindrical Tank System
12.1 Introduction
12.2 Methodology
12.2.1 Mathematical Model of Cylindrical Tank
12.2.2 Description of Metaheuristic Techniques
12.2.2.1 Flower Pollination Algorithm (FPA)
12.2.2.2 Bacterial Foraging Optimization Algorithm (BFOA)
12.3 Results and Discussion
12.4 Conclusion
References
Chapter 13: A Hybrid Algorithm Based on CSO and PSO for Parametric Optimization of Liquid Flow Model
13.1 Introduction
13.2 Experimental Setup Liquid Flow Control Process
13.3 Modeling of the Liquid Flow Process
13.4 Proposed Methodology
13.4.1 Hybrid GAPSO
13.4.2 Parameters Setting
13.5 Performance Analysis
13.5.1 Computational Efficiency Test
13.5.2 Convergence Speed
13.5.3 Accuracy Test
13.6 Finding Optimal Condition for Liquid Flow
13.7 Conclusions
References
Chapter 14: Modeling of Improved Deep Learning Algorithm for Detection of Type 2 Diabetes
14.1 Introduction
14.2 Methodology
14.2.1 Datasets
14.2.2 Imbalanced Datasets
14.2.3 Synthetic Minority Over-Sampling Technique (SMOTE)
14.3 Proposed Flow Diagram
14.4 Deep Neural Network for Data Classification
14.5 Experimental Result Analysis
14.5.1 Performance Measure
14.5.2 Comparison with Existing System
14.6 Conclusions
References
Chapter 15: Human Activity Recognition (HAR), Prediction, and Analysis Using Machine Learning
15.1 Introduction
15.2 Related Works
15.3 Proposed Method for Human Action Recognition
15.3.1 Data Collection Overview
15.3.2 Signal Processing
15.3.3 Feature Selection
15.3.4 Exploratory Data Analysis
15.3.5 Data Preprocessing
15.3.6 Exploratory Data Analysis for Static and Dynamic Activities
15.3.7 Visualizing Data Using t-SNE
15.4 Machine Learning Algorithm
15.4.1 Logistics Regression
15.4.2 Random Forest
15.4.3 Decision Tree
15.4.4 Support Vector Machine
15.4.5 K Nearest Neighbor (KNN)
15.4.6 Naïve Bayes
15.4.7 Data Preprocessing
15.5 Experimental Results
15.6 Conclusion
References
Index