Explainable Machine Learning Models and Architectures

Suman Lata Tripathi, PhD, is a professor at Lovely Professional University with more than 21 years of experience in academics. She has published more than 103 research papers in refereed journals and conferences. She has organized several workshops, summer internships, and expert lectures for students, and she has worked as a session chair, conference steering committee member, editorial board member, and reviewer for IEEE journals and conferences. She has published three books and currently has multiple volumes scheduled for publication from Wiley-Scrivener. Mufti Mahmud, PhD, is an associate professor of cognitive computing at the Department of Computer Science of Nottingham Trent University, UK. He is the Coordinator of the Computer Science and Informatics Unit of Assessment of Research Excellence Framework at NTU and the deputy group leader of the Interactive Systems Research Group and the Cognitive Computing & Brain Informatics research group. He is also an active member of the Computing and Informatics Research Centre and the Medical Technologies Innovation Facility. He is a member of numerous societies and research committees.

• Preface xiiiAcknowledgements xv1 A Comprehensive Review of Various Machine Learning Techniques 1Pooja Pathak and Parul Choudhary1.1 Introduction 11.1.1 Random Forest 21.1.2 Decision Tree 31.1.3 Support Vector Machine 41.1.4 Naive Bayes 51.1.5 K-Means Clustering 61.1.6 Principal Component Analysis 61.1.7 Linear Regression 61.1.8 Logistic Regression 71.1.9 Semi-Supervised Learning 81.1.10 Transductive SVM 91.1.11 Generative Models 91.1.12 Self-Training 91.1.13 Relearning 91.2 Conclusions 92 Artificial Intelligence and Image Recognition Algorithms 11Siddharth, Anuranjana and Sanmukh Kaur2.1 Introduction 122.2 Traditional Image Recognition Algorithms 132.2.1 Harris Corner Detector (1988) 132.2.2 SIFT (2004) 152.2.3 ASIFT 162.2.4 SURF (2006) 172.3 Neural Network-Based Algorithms 212.4 Convolutional Neural Network Architecture 222.5 Various CNN Architectures 232.5.1 LeNet-5 (1998) 232.5.2 AlexNet (2012) 242.5.3 VGGNet (2014) 242.5.4 GoogleNet (2015) 243 Efficient Architectures and Trade-Offs for FPGA-Based Real-Time Systems 31L.M.I. Leo Joseph, J. Ajayan, Sandip Bhattacharya and Sreedhar Kollem3.1 Overview of FPGA-Based Real-Time System 313.1.1 Key Elements of Real-Time System 323.1.2 Real-Time System and its Computation 323.1.3 FPGA Functionality and Applications 333.1.4 FPGA Applications 333.1.5 FPGA Architecture 343.1.6 Reconfigurable Architectures 353.2 Hybrid FPGA Configurations and its Algorithms 383.2.1 Hybrid FPGA 383.2.2 Hybrid FPGA Architecture 393.2.3 Hybrid FPGA Configuration 403.3 Hybrid FPGA Algorithms 423.3.1 Relevance of Hardware-Accelerated Architecture to FPGA Software Implementation 443.4 CNN Hardware Accelerator Architecture Overview 463.5 Summary 474 A Low-Power Audio Processing Using Machine Learning Module on FPGA and Applications 49Suman Lata Tripathi, Dasari Lakshmi Prasanna and Mufti Mahmud4.1 Introduction 494.2 Existing Machine Learning Modules and Audio Classifiers 504.3 Audio Processing Module Using Machine Learning 564.4 Application of Proposed FPGA-Based ML Models 574.5 Implementation of a Microphone on FPGA 594.6 Conclusion 604.7 Future Scope 605 Synthesis and Time Analysis of FPGA-Based DIT-FFT Module for Efficient VLSI Signal Processing Applications 65Siba Kumar Panda, Konasagar Achyut and Dhruba Charan Panda5.1 Introduction 665.2 Implementation of DIT-FFT Algorithm 675.2.1 A Quick Overview of DIT-FFT 675.2.2 Algorithmic Representation with Example 695.2.3 Simulated Output Waveform 695.3 Synthesis of Designed Circuit 715.4 Static Timing Analysis of Designed Circuit 735.5 Result and Discussion 775.6 Conclusion 776 Artificial Intelligence-Based Active Virtual Voice Assistant 81Swathi Gowroju, G. Mounika, D. Bhavana, Shaik Abdul Latheef and A. Abhilash6.1 Introduction 826.2 Literature Survey 836.3 System Functions 876.4 Model Training 886.5 Discussion 906.5.1 Furnishing Movie Recommendations 916.5.2 KNN Algorithm Book Recommendation 926.6 Results 936.7 Conclusion 1027 Image Forgery Detection: An Approach with Machine Learning 105Madhusmita Mishra, Silvia Tittotto and Santos Kumar Das7.1 Introduction 1057.2 Historical Background 1077.3 CNN Architecture 1097.4 Analysis of Error Level of Image 1137.5 Proposed Model of Image Forgery Detection, Results and Discussion 1157.6 Conclusion 1187.7 Future Research Directions 1198 Applications of Artificial Neural Networks in Optical Performance Monitoring 123Isra Imtiyaz, Anuranjana, Sanmukh Kaur and Anubhav Gautam8.1 Introduction 1238.2 Algorithms Employed for Performance Monitoring 1298.2.1 Artificial Neural Networks 1298.2.2 Deep Neural Networks 1308.2.3 Convolutional Neural Networks 1318.2.3.1 Convolutional Layer 1318.2.3.2 Non-Linear Layer 1328.2.3.3 Pooling Layer 1328.2.3.4 Fully Connected Layer 1328.2.4 Support Vector Regression (SVR) 1338.2.5 Support Vector Machine (SVM) 1338.2.6 Kernel Ridge Regression (KRR) 1338.2.7 Long Short-Term Memory (LSTM) 1338.3 Artificial Intelligence (AI) Methods, Performance Monitoring and Applications in Optical Networks 1348.3.1 Performance Monitoring 1348.3.2 Applications of AI in Optical Networking 1358.4 Optical Impairments and Fault Management 1358.4.1 Noise 1358.4.2 Distortion 1358.4.3 Timing 1368.4.4 Component Faults 1368.4.5 Transmission Impairments 1378.4.6 Fault Management in Optical Network 1378.5 Conclusion 1389 Website Development with Django Web Framework 141Sanmukh Kaur, Anuranjana and Yashasvi Roy9.1 Introduction 1419.2 Salient Features of Django 1429.2.1 Complete 1429.2.2 Versatile 1429.2.3 Secure 1429.2.4 Scalable 1439.2.5 Maintainable 1439.2.6 Portable 1439.3 UI Design 1439.3.1 HTML 1439.3.2 CSS 1449.3.3 Bootstrap 1449.4 Methodology 1449.5 UI Design 1449.6 Backend Development 1489.6.1 Login Page 1489.6.2 Registration Page 1499.6.3 User Tracking 1499.7 Ouputs 1509.8 Conclusion 15210 Revenue Forecasting Using Machine Learning Models 155Yashasvi Roy and Sanmukh Kaur10.1 Introduction 15510.2 Types of Forecasting 15610.2.1 Qualitative Forecasting 15610.2.1.1 Industries That Use Qualitative Forecasting 15710.2.1.2 Qualitative Forecasting Methods 15810.2.2 Quantitative Forecasting 15810.2.2.1 Quantitative Forecasting Methods 15910.2.3 Artificial Intelligence Forecasting 16010.2.3.1 Artificial Neural Network (ANN) 16010.2.3.2 Support Vector Machine (SVM) 16110.3 Types of ML Models Used in Finance 16210.3.1 Linear Regression 16210.3.1.1 Simple Linear Regression 16210.3.1.2 Multiple Linear Regression 16210.3.2 Ridge Regression 16310.3.3 Decision Tree 16410.3.3.1 Prediction of Continuous Variables 16410.3.3.2 Prediction of Categorical Variables 16510.3.4 Random Forest Regressor 16510.3.5 Gradient Boosting Regression 16610.3.5.1 Advantages of Gradient Boosting 16710.4 Model Performance 16710.4.1 R-Squared Method 16710.4.2 Mean Squared Error (MSE) 16710.4.3 Root Mean Square Error (RMSE) 16810.5 Conclusion 16811 Application of Machine Learning Optimization Techniques in Wind Resource Assessment 171Udhayakumar K. and Krishnamoorthy R.11.1 Introduction 17211.2 Wind Data Analysis Methods 17311.2.1 Wind Characteristics Parameters 17311.2.2 Wind Speed Distribution Methods 17311.2.3 Weibull Method 17411.2.4 Goodness of Fit 17511.3 Wind Site and Measurement Details 17511.3.1 Seasonal Wind Periods 17611.3.2 Machine Learning and Optimization Techniques 17611.3.2.1 Moth Flame Optimization (MFO) Method 17611.4 Results and Discussions 18011.4.1 Wind Characteristics 18211.4.1.1 Kayathar Station (Onshore) 18211.4.1.2 Gulf of Khambhat (Gujarat Offshore) Station 18711.4.1.3 Jafrabad (Gujarat-Nearshore) 19211.4.2 Wind Distribution Fitting 19511.4.2.1 Kayathar Station (Onshore) 19611.4.2.2 Bimodal Behaviour 19611.4.2.3 Gulf of Khambhat (Offshore) Wind Distribution 20211.4.2.4 Jafrabad Station (Nearshore) Distribution Fitting 20311.4.3 Optimization Methods for Parameter Estimation 21211.4.3.1 Optimization Parameters Comparison 21211.4.4 Wind Power Density Analysis (WPD) 21411.4.4.1 Comparison of Wind Power Density 21511.5 Research Summary 22111.6 Conclusions 22212 IoT to Scale-Up Smart Infrastructure in Indian Cities: A New Paradigm 227Indu Bala, Simarpreet Kaur, Lavpreet Kaur and Pavan Thimmavajjala12.1 Introduction 22812.2 Technological Progress: A Brief History 22912.3 What is the Internet of Things (IoT)? 23012.4 Economic Effects of Internet of Things 23012.5 Infrastructure and Smart Infrastructure: The Difference 23212.5.1 What is Smart Infrastructure? 23312.5.2 What are the Principles of Smart Infrastructure? 23412.5.3 Components of IoT-Based Smart City Project 23512.6 Architecture for Smart Cities 23612.6.1 Networking Technologies 23712.6.2 Network Topologies 23712.6.3 Network Architectures 23812.6.3.1 Home Area Networks (HANs) 23812.6.3.2 Field/Neighborhood Area Networks (FANs/NANs) 23812.6.3.3 Wide Area Networks (WANs) 23812.6.3.4 Network Protocols 23812.7 IoT Technology in India’s Smart Cities: The Current Scenario 23912.8 Challenges in IoT-Based Smart City Projects 24312.8.1 Technological Challenges 24312.8.1.1 Privacy and Security 24312.8.1.2 Smart Sensors and Infrastructure Essentials 24312.8.1.3 Networking in IoT Systems 24412.8.1.4 Big Data Analytics 24412.8.2 Financial - Economic Challenges 24412.9 Role of Explainable AI 24512.10 Conclusion and Future Scope 246References 246Index 251