Advanced Analytics and Deep Learning Models

Archana Mire, PhD, is an assistant professor in the Computer Engineering Department, Terna Engineering College, Navi Mumbai, India. She has published many research articles in peer-reviewed journals. Shaveta Malik, PhD, is an associate professor in the Computer Engineering Department (NBA accredited), Terna Engineering College, Nerul, India. She has published many research articles in peer-reviewed journals. Amit Kumar Tyagi, PhD, is an assistant professor and senior researcher at Vellore Institute of Technology (VIT), Chennai Campus, India. He received his PhD in 2018 from Pondicherry Central University, India. He has published more than 8 patents in the area of deep learning, Internet of Things, cyber-physical systems, and computer vision.

• Preface xixPart 1: Introduction to Computer Vision 11 Artificial Intelligence in Language Learning: Practices and Prospects 3Khushboo Kuddus1.1 Introduction 41.2 Evolution of CALL 51.3 Defining Artificial Intelligence 71.4 Historical Overview of AI in Education and Language Learning 71.5 Implication of Artificial Intelligence in Education 81.5.1 Machine Translation 91.5.2 Chatbots 91.5.3 Automatic Speech Recognition Tools 91.5.4 Autocorrect/Automatic Text Evaluator 111.5.5 Vocabulary Training Applications 121.5.6 Google Docs Speech Recognition 121.5.7 Language MuseTM Activity Palette 131.6 Artificial Intelligence Tools Enhance the Teaching and Learning Processes 131.6.1 Autonomous Learning 131.6.2 Produce Smart Content 131.6.3 Task Automation 131.6.4 Access to Education for Students with Physical Disabilities 141.7 Conclusion 14References 152 Real Estate Price Prediction Using Machine Learning Algorithms 19Palak Furia and Anand Khandare2.1 Introduction 202.2 Literature Review 202.3 Proposed Work 212.3.1 Methodology 212.3.2 Work Flow 222.3.3 The Dataset 222.3.4 Data Handling 232.3.4.1 Missing Values and Data Cleaning 232.3.4.2 Feature Engineering 242.3.4.3 Removing Outliers 252.4 Algorithms 272.4.1 Linear Regression 272.4.2 LASSO Regression 272.4.3 Decision Tree 282.4.4 Support Vector Machine 282.4.5 Random Forest Regressor 282.4.6 XGBoost 292.5 Evaluation Metrics 292.6 Result of Prediction 30References 313 Multi-Criteria–Based Entertainment Recommender System Using Clustering Approach 33Chandramouli Das, Abhaya Kumar Sahoo and Chittaranjan Pradhan3.1 Introduction 343.2 Work Related Multi-Criteria Recommender System 353.3 Working Principle 383.3.1 Modeling Phase 393.3.2 Prediction Phase 393.3.3 Recommendation Phase 403.3.4 Content-Based Approach 403.3.5 Collaborative Filtering Approach 413.3.6 Knowledge-Based Filtering Approach 413.4 Comparison Among Different Methods 423.4.1 MCRS Exploiting Aspect-Based Sentiment Analysis 423.4.1.1 Discussion and Result 433.4.2 User Preference Learning in Multi-Criteria Recommendation Using Stacked Autoencoders by Tallapally et al. 463.4.2.1 Dataset and Evaluation Matrix 463.4.2.2 Training Setting 493.4.2.3 Result 493.4.3 Situation-Aware Multi-Criteria Recommender System: Using Criteria Preferences as Contexts by Zheng 493.4.3.1 Evaluation Setting 503.4.3.2 Experimental Result 503.4.4 Utility-Based Multi-Criteria Recommender Systems by Zheng 513.4.4.1 Experimental Dataset 513.4.4.2 Experimental Result 523.4.5 Multi-Criteria Clustering Approach by Wasid and Ali 533.4.5.1 Experimental Evaluation 533.4.5.2 Result and Analysis 533.5 Advantages of Multi-Criteria Recommender System 543.5.1 Revenue 573.5.2 Customer Satisfaction 573.5.3 Personalization 573.5.4 Discovery 583.5.5 Provide Reports 583.6 Challenges of Multi-Criteria Recommender System 583.6.1 Cold Start Problem 583.6.2 Sparsity Problem 593.6.3 Scalability 593.6.4 Over Specialization Problem 593.6.5 Diversity 593.6.6 Serendipity 593.6.7 Privacy 603.6.8 Shilling Attacks 603.6.9 Gray Sheep 603.7 Conclusion 60References 614 Adoption of Machine/Deep Learning in Cloud With a Case Study on Discernment of Cervical Cancer65Jyothi A. P., S. Usha and Archana H. R.4.1 Introduction 664.2 Background Study 694.3 Overview of Machine Learning/Deep Learning 724.4 Connection Between Machine Learning/Deep Learning and Cloud Computing 744.5 Machine Learning/Deep Learning Algorithm 744.5.1 Supervised Learning 744.5.2 Unsupervised Learning 774.5.3 Reinforcement or Semi-Supervised Learning 774.5.3.1 Outline of ML Algorithms 774.6 A Project Implementation on Discernment of Cervical Cancer by Using Machine/Deep Learning in Cloud 934.6.1 Proposed Work 944.6.1.1 MRI Dataset 944.6.1.2 Pre Processing 954.6.1.3 Feature Extraction 964.6.2 Design Methodology and Implementation 974.6.3 Results 1004.7 Applications 1014.7.1 Cognitive Cloud 1024.7.2 Chatbots and Smart Personal Assistants 1034.7.3 IoT Cloud 1034.7.4 Business Intelligence 1034.7.5 AI-as-a-Service 1044.8 Advantages of Adoption of Cloud in Machine Learning/ Deep Learning 1044.9 Conclusion 105References 1065 Machine Learning and Internet of Things–Based Models for Healthcare Monitoring 111Shruti Kute, Amit Kumar Tyagi, Aswathy S.U. and Shaveta Malik5.1 Introduction 1125.2 Literature Survey 1135.3 Interpretable Machine Learning in Healthcare 1145.4 Opportunities in Machine Learning for Healthcare 1165.5 Why Combining IoT and ML? 1195.5.1 ML-IoT Models for Healthcare Monitoring 1195.6 Applications of Machine Learning in Medical and Pharma 1215.7 Challenges and Future Research Direction 1225.8 Conclusion 123References 1236 Machine Learning–Based Disease Diagnosis and Prediction for E-Healthcare System 127Shruti Suhas Kute, Shreyas Madhav A. V., Shabnam Kumari and Aswathy S. U.6.1 Introduction 1286.2 Literature Survey 1296.3 Machine Learning Applications in Biomedical Imaging 1326.4 Brain Tumor Classification Using Machine Learning and IoT 1346.5 Early Detection of Dementia Disease Using Machine Learning and IoT-Based Applications 1356.6 IoT and Machine Learning-Based Diseases Prediction and Diagnosis System for EHRs 1376.7 Machine Learning Applications for a Real-Time Monitoring of Arrhythmia Patients Using IoT 1406.8 IoT and Machine Learning–Based System for Medical Data Mining 1416.9 Conclusion and Future Works 143References 144Part 2: Introduction to Deep Learning and its Models 1497 Deep Learning Methods for Data Science 151K. Indira, Kusumika Krori Dutta, S. Poornima and Sunny Arokia Swamy Bellary7.1 Introduction 1527.2 Convolutional Neural Network 1527.2.1 Architecture 1547.2.2 Implementation of CNN 1547.2.3 Simulation Results 1577.2.4 Merits and Demerits 1587.2.5 Applications 1597.3 Recurrent Neural Network 1597.3.1 Architecture 1607.3.2 Types of Recurrent Neural Networks 1617.3.2.1 Simple Recurrent Neural Networks 1617.3.2.2 Long Short-Term Memory Networks 1627.3.2.3 Gated Recurrent Units (GRUs) 1647.3.3 Merits and Demerits 1677.3.3.1 Merits 1677.3.3.2 Demerits 1677.3.4 Applications 1677.4 Denoising Autoencoder 1687.4.1 Architecture 1697.4.2 Merits and Demerits 1697.4.3 Applications 1707.5 Recursive Neural Network (RCNN) 1707.5.1 Architecture 1707.5.2 Merits and Demerits 1727.5.3 Applications 1727.6 Deep Reinforcement Learning 1737.6.1 Architecture 1747.6.2 Merits and Demerits 1747.6.3 Applications 1747.7 Deep Belief Networks (DBNS) 1757.7.1 Architecture 1767.7.2 Merits and Demerits 1767.7.3 Applications 1767.8 Conclusion 177References 1778 A Proposed LSTM-Based Neuromarketing Model for Consumer Emotional State Evaluation Using EEG 181Rupali Gill and Jaiteg Singh8.1 Introduction 1828.2 Background and Motivation 1838.2.1 Emotion Model 1838.2.2 Neuromarketing and BCI 1848.2.3 EEG Signal 1858.3 Related Work 1858.3.1 Machine Learning 1868.3.2 Deep Learning 1918.3.2.1 Fast Feed Neural Networks 1938.3.2.2 Recurrent Neural Networks 1938.3.2.3 Convolutional Neural Networks 1948.4 Methodology of Proposed System 1958.4.1 DEAP Dataset 1968.4.2 Analyzing the Dataset 1968.4.3 Long Short-Term Memory 1978.4.4 Experimental Setup 1978.4.5 Data Set Collection 1978.5 Results and Discussions 1988.5.1 LSTM Model Training and Accuracy 1988.6 Conclusion 199References 1999 An Extensive Survey of Applications of Advanced Deep Learning Algorithms on Detection of Neurodegenerative Diseases and the Tackling Procedure in Their Treatment Protocol 207Vignesh Baalaji S., Vergin Raja Sarobin M., L. Jani Anbarasi, Graceline Jasmine S. and Rukmani P.9.1 Introduction 2089.2 Story of Alzheimer’s Disease 2089.3 Datasets 2109.3.1 ADNI 2109.3.2 OASIS 2109.4 Story of Parkinson’s Disease 2119.5 A Review on Learning Algorithms 2129.5.1 Convolutional Neural Network (CNN) 2129.5.2 Restricted Boltzmann Machine 2139.5.3 Siamese Neural Networks 2139.5.4 Residual Network (ResNet) 2149.5.5 U-Net 2149.5.6 LSTM 2149.5.7 Support Vector Machine 2159.6 A Review on Methodologies 2159.6.1 Prediction of Alzheimer’s Disease 2159.6.2 Prediction of Parkinson’s Disease 2219.6.3 Detection of Attacks on Deep Brain Stimulation 2239.7 Results and Discussion 2249.8 Conclusion 224References 22710 Emerging Innovations in the Near Future Using Deep Learning Techniques 231Akshara Pramod, Harsh Sankar Naicker and Amit Kumar Tyagi10.1 Introduction 23210.2 Related Work 23410.3 Motivation 23510.4 Future With Deep Learning/Emerging Innovations in Near Future With Deep Learning 23610.4.1 Deep Learning for Image Classification and Processing 23710.4.2 Deep Learning for Medical Image Recognition 23710.4.3 Computational Intelligence for Facial Recognition 23810.4.4 Deep Learning for Clinical and Health Informatics 23810.4.5 Fuzzy Logic for Medical Applications 23910.4.6 Other Intelligent-Based Methods for Biomedical and Healthcare 23910.4.7 Other Applications 23910.5 Open Issues and Future Research Directions 24410.5.1 Joint Representation Learning From User and Item Content Information 24410.5.2 Explainable Recommendation With Deep Learning 24510.5.3 Going Deeper for Recommendation 24510.5.4 Machine Reasoning for Recommendation 24610.5.5 Cross Domain Recommendation With Deep Neural Networks 24610.5.6 Deep Multi-Task Learning for Recommendation 24710.5.7 Scalability of Deep Neural Networks for Recommendation 24710.5.8 Urge for a Better and Unified Evaluation 24810.6 Deep Learning: Opportunities and Challenges 24910.7 Argument with Machine Learning and Other Available Techniques 25010.8 Conclusion With Future Work 251Acknowledgement 252References 25211 Optimization Techniques in Deep Learning Scenarios: An Empirical Comparison 255Ajeet K. Jain, PVRD Prasad Rao and K. Venkatesh Sharma11.1 Introduction 25611.1.1 Background and Related Work 25611.2 Optimization and Role of Optimizer in DL 25811.2.1 Deep Network Architecture 25911.2.2 Proper Initialization 26011.2.3 Representation, Optimization, and Generalization 26111.2.4 Optimization Issues 26111.2.5 Stochastic GD Optimization 26211.2.6 Stochastic Gradient Descent with Momentum 26311.2.7 SGD With Nesterov Momentum 26411.3 Various Optimizers in DL Practitioner Scenario 26511.3.1 AdaGrad Optimizer 26511.3.2 RMSProp 26711.3.3 Adam 26711.3.4 AdaMax 26911.3.5 AMSGrad 26911.4 Recent Optimizers in the Pipeline 27011.4.1 EVE 27011.4.2 RAdam 27111.4.3 MAS (Mixing ADAM and SGD) 27111.4.4 Lottery Ticket Hypothesis 27211.5 Experiment and Results 27311.5.1 Web Resource 27311.5.2 Resource 27711.6 Discussion and Conclusion 278References 279Part 3: Introduction to Advanced Analytics 28312 Big Data Platforms 285Sharmila Gaikwad and Jignesh Patil12.1 Visualization in Big Data 28612.1.1 Introduction to Big Data 28612.1.2 Techniques of Visualization 28712.1.3 Case Study on Data Visualization 30212.2 Security in Big Data 30512.2.1 Introduction of Data Breach 30512.2.2 Data Security Challenges 30612.2.3 Data Breaches 30712.2.4 Data Security Achieved 30712.2.5 Findings: Case Study of Data Breach 30912.3 Conclusion 309References 30913 Smart City Governance Using Big Data Technologies 311K. Raghava Rao and D. Sateesh Kumar13.1 Objective 31213.2 Introduction 31213.3 Literature Survey 31413.4 Smart Governance Status 31413.4.1 International 31413.4.2 National 31613.5 Methodology and Implementation Approach 31813.5.1 Data Generation 31913.5.2 Data Acquisition 31913.5.3 Data Analytics 31913.6 Outcome of the Smart Governance 32213.7 Conclusion 323References 32314 Big Data Analytics With Cloud, Fog, and Edge Computing 325Deepti Goyal, Amit Kumar Tyagi and Aswathy S. U.14.1 Introduction to Cloud, Fog, and Edge Computing 32614.2 Evolution of Computing Terms and Its Related Works 33014.3 Motivation 33214.4 Importance of Cloud, Fog, and Edge Computing in Various Applications 33314.5 Requirement and Importance of Analytics (General) in Cloud, Fog, and Edge Computing 33414.6 Existing Tools for Making a Reliable Communication and Discussion of a Use Case (with Respect to Cloud, Fog, and Edge Computing) 33514.6.1 CloudSim 33514.6.2 SPECI 33614.6.3 Green Cloud 33614.6.4 OCT (Open Cloud Testbed) 33714.6.5 Open Cirrus 33714.6.6 GroudSim 33814.6.7 Network CloudSim 33814.7 Tools Available for Advanced Analytics (for Big Data Stored in Cloud, Fog, and Edge Computing Environment) 33814.7.1 Microsoft HDInsight 33814.7.2 Skytree 33914.7.3 Splice Machine 33914.7.4 Spark 33914.7.5 Apache SAMOA 33914.7.6 Elastic Search 33914.7.7 R-Programming 33914.8 Importance of Big Data Analytics for Cyber-Security and Privacy for Cloud-IoT Systems 34014.8.1 Risk Management 34014.8.2 Predictive Models 34014.8.3 Secure With Penetration Testing 34014.8.4 Bottom Line 34114.8.5 Others: Internet of Things-Based Intelligent Applications 34114.9 An Use Case with Real World Applications (with Respect to Big Data Analytics) Related to Cloud, Fog, and Edge Computing 34114.10 Issues and Challenges Faced by Big Data Analytics (in Cloud, Fog, and Edge Computing Environments) 34214.10.1 Cloud Issues 34314.11 Opportunities for the Future in Cloud, Fog, and Edge Computing Environments (or Research Gaps) 34414.12 Conclusion 345References 34615 Big Data in Healthcare: Applications and Challenges 351V. Shyamala Susan, K. Juliana Gnana Selvi and Ir. Bambang Sugiyono Agus Purwono15.1 Introduction 35215.1.1 Big Data in Healthcare 35215.1.2 The 5V’s Healthcare Big Data Characteristics 35315.1.2.1 Volume 35315.1.2.2 Velocity 35315.1.2.3 Variety 35315.1.2.4 Veracity 35315.1.2.5 Value 35315.1.3 Various Varieties of Big Data Analytical (BDA) in Healthcare 35315.1.4 Application of Big Data Analytics in Healthcare 35415.1.5 Benefits of Big Data in the Health Industry 35515.2 Analytical Techniques for Big Data in Healthcare 35615.2.1 Platforms and Tools for Healthcare Data 35715.3 Challenges 35715.3.1 Storage Challenges 35715.3.2 Cleaning 35815.3.3 Data Quality 35815.3.4 Data Security 35815.3.5 Missing or Incomplete Data 35815.3.6 Information Sharing 35815.3.7 Overcoming the Big Data Talent and Cost Limitations 35915.3.8 Financial Obstructions 35915.3.9 Volume 35915.3.10 Technology Adoption 36015.4 What is the Eventual Fate of Big Data in Healthcare Services? 36015.5 Conclusion 361References 36116 The Fog/Edge Computing: Challenges, Serious Concerns, and the Road Ahead 365Varsha. R., Siddharth M. Nair and Amit Kumar Tyagi16.1 Introduction 36616.1.1 Organization of the Work 36816.2 Motivation 36816.3 Background 36916.4 Fog and Edge Computing–Based Applications 37116.5 Machine Learning and Internet of Things–Based Cloud, Fog, and Edge Computing Applications 37416.6 Threats Mitigated in Fog and Edge Computing–Based Applications 37616.7 Critical Challenges and Serious Concerns Toward Fog/Edge Computing and Its Applications 37816.8 Possible Countermeasures 38116.9 Opportunities for 21st Century Toward Fog and Edge Computing 38316.9.1 5G and Edge Computing as Vehicles for Transformation of Mobility in Smart Cities 38316.9.2 Artificial Intelligence for Cloud Computing and Edge Computing 38416.10 Conclusion 387References 387Index 391