Data Science and Data Analytics: Opportunities and Challenges
- Length: 482 pages
- Edition: 1
- Language: English
- Publisher: Chapman and Hall/CRC
- Publication Date: 2021-09-23
- ISBN-10: 0367628821
- ISBN-13: 9780367628826
- Sales Rank: #0 (See Top 100 Books)
Data science is a multi-disciplinary field that uses scientific methods, processes, algorithms, and systems to extract knowledge and insights from structured (labeled) and unstructured (unlabeled) data. It is the future of Artificial Intelligence (AI) and a necessity of the future to make things easier and more productive. In simple terms, data science is the discovery of data or uncovering hidden patterns (such as complex behaviors, trends, and inferences) from data. Moreover, Big Data analytics/data analytics are the analysis mechanisms used in data science by data scientists. Several tools, such as Hadoop, R, etc., are used to analyze this large amount of data to predict valuable information and for decision-making. Note that structured data can be easily analyzed by efficient (available) business intelligence tools, while most of the data (80% of data by 2020) is in an unstructured form that requires advanced analytics tools. But while analyzing this data, we face several concerns, such as complexity, scalability, privacy leaks, and trust issues.
Data science helps us to extract meaningful information or insights from unstructured or complex or large amounts of data (available or stored virtually in the cloud). Data Science and Data Analytics: Opportunities and Challenges covers all possible areas, applications with arising serious concerns, and challenges in this emerging field in detail with a comparative analysis/taxonomy.
FEATURES
- Gives the concept of data science, tools, and algorithms that exist for many useful applications
- Provides many challenges and opportunities in data science and data analytics that help researchers to identify research gaps or problems
- Identifies many areas and uses of data science in the smart era
- Applies data science to agriculture, healthcare, graph mining, education, security, etc.
Academicians, data scientists, and stockbrokers from industry/business will find this book useful for designing optimal strategies to enhance their firm’s productivity.
Cover Half Title Title Page Copyright Page Contents Preface Editor Contributors Section I: Introduction about Data Science and Data Analytics 1. Data Science and Data Analytics: Artificial Intelligence and Machine Learning Integrated Based Approach 1.1 Introduction 1.2 Artificial Intelligence 1.3 Machine Learning (ML) 1.3.1 Regression 1.3.1.1 Linear Regression 1.3.1.2 Logistic Regression Multi-class Logistic Regression Polytomous Logistic Regression 1.3.2 Support Vector Machine (SVM) 1.4 Deep Learning (DL) 1.4.1 Methods for Deep Learning 1.4.1.1 Convolutional Neural Networks (CNNs) General Model of Convolutional Neural Network Convolution Layer Nonlinear Activation Function Pooling Layer Fully Connected Layer Last Layer Activation Function 1.4.1.2 Extreme Learning Machine 1.4.1.3 Transfer Learning (TL) Important Considerations for Transfer Learning Types of Transfer Learning 1.5 Bio-inspired Algorithms for Data Analytics 1.6 Conclusion References 2. IoT Analytics/Data Science for IoT 2.1 Preface 2.1.1 Data Science Components 2.1.2 Method for Data Science 2.1.3 The Internet of Stuff 2.1.3.1 Difficulties in the Comprehension of Stuff on the Internet 2.1.3.2 Sub-domain of Data Science for IoT 2.1.3.3 IoT and Relationship with Data 2.1.3.4 IoT Applications in Data Science Challenges 2.1.3.5 Ways to Distribute Algorithms in Computer Science to IoT Data 2.2 Computational Methodology-IoT Science for Data Science 2.2.1 Regression 2.2.2 Set of Trainings 2.2.3 Pre-processing 2.2.4 Sensor Fusion Leverage for the Internet of Things 2.3 Methodology-IoT Mechanism of Privacy 2.3.1 Principles for IoT Security 2.3.2 IoT Architecture Offline 2.3.3 Offline IoT Architecture 2.3.4 Online IoT Architecture 2.3.5 IoT Security Issues 2.3.6 Applications 2.4 Consummation References 3. A Model to Identify Agriculture Production Using Data Science Techniques 3.1 Agriculture System Application Based on GPS/GIS Gathered Information 3.1.1 Important Tools Required for Developing GIS/GPS-Based Agricultural System 3.1.1.1 Information (Gathered Data) 3.1.1.2 Map 3.1.1.3 System Apps 3.1.1.4 Data Analysis 3.1.2 GPS/GIS in Agricultural Conditions 3.1.2.1 GIS System in Agriculture 3.1.3 System Development Using GIS and GPS Data 3.2 Design of Interface to Extract Soil Moisture and Mineral Content in Agricultural Lands 3.2.1 Estimating Level of Soil Moisture and Mineral Content Using COSMIC-RAY (C-RAY) Sensing Technique 3.2.1.1 Cosmic 3.2.2 Soil Moisture and Mineral Content Measurement Using Long Duration Optical Fiber Grating (LDOPG) 3.2.3 Moisture Level and Mineral Content Detection System Using a Sensor Device 3.2.4 Soil Moisture Experiment 3.2.4.1 Dataset Description 3.2.5 Experimental Result 3.3 Analysis and Guidelines for Seed Spacing 3.3.1 Correct Spacing 3.3.2 System Components 3.3.2.1 Electronic Compass 3.3.2.2 Optical Flow Sensor 3.3.2.3 Motor Driver 3.3.2.4 Microcontroller 3.4 Analysis of Spread of Fertilizers 3.4.1 Relationship between Soil pH Value and Nutrient Availability 3.4.2 Methodology 3.4.2.1 Understand Define Phase 3.4.2.2 Analysis and Quick Design Phase 3.4.2.3 Prototype Development Phase 3.4.2.4 Testing Phase 3.4.3 System Architecture 3.4.4 Experimental Setup 3.4.5 Implementation Phase 3.4.6 Experimental Results 3.5 Conclusion and Future Work References 4. Identification and Classification of Paddy Crop Diseases Using Big Data Machine Learning Techniques 4.1 Introduction 4.1.1 Overview of Paddy Crop Diseases 4.1.2 Overview of Big Data 4.1.2.1 Features of Big Data 4.1.3 Overview of Machine Learning Techniques 4.1.3.1 K-Nearest Neighbor 4.1.3.2 Support Vector Machine 4.1.3.3 K-Means 4.1.3.4 Fuzzy C-Means 4.1.3.5 Decision Tree 4.1.4 Overview of Big Data Machine Learning Tools 4.1.4.1 Hadoop 4.1.4.2 Hadoop Distributed File System (HDFS) 4.1.4.3 YARN ("Yet Another Resource Negotiator") 4.2 Related Work 4.2.1 Image Recognition/Processing 4.2.2 Classification and Feature Extraction 4.2.3 Problems and Diseases 4.3 Proposed Architecture 4.3.1 Image Acquisition 4.3.2 Image Enhancement 4.3.3 Image Segmentation 4.3.4 Feature Extraction 4.3.5 Classification 4.4 Proposed Algorithms and Implementation Details 4.4.1 Image Preprocessing 4.4.2 Image Segmentation and the Fuzzy C-Means Model Using Spark 4.4.3 Feature Extraction 4.4.4 Classification 4.4.4.1 Support Vector Machine (SVM) 4.4.4.2 Naïve Bayes 4.4.4.3 Decision Tree and Random Forest 4.5 Result Analysis 4.5.1 Comparison of Speed-up Performance between the Spark-Based and Hadoop-Based FCM Approach 4.5.2 Comparison of Scale-up Performance between the Spark-Based and Hadoop-Based FCM Approach 4.5.3 Result Analysis of Various Segmentation Techniques 4.5.4 Results of Disease Identification 4.6 Conclusion and Future Work References Section II: Algorithms, Methods, and Tools for Data Science and Data Analytics 5. Crop Models and Decision Support Systems Using Machine Learning 5.1 Introduction 5.1.1 Decision Support System 5.1.2 Decision Support System for Crop Yield 5.1.3 What Is Crop Modeling? 5.1.4 Necessity of Crop Modeling 5.1.5 Recent Trends in Crop Modeling 5.2 Methodologies 5.2.1 Machine-Learning-Based Techniques 5.2.2 Deep-Learning-Based Techniques 5.2.3 Hyper-Spectral Imaging 5.2.4 Popular Band Selection Techniques 5.2.5 Leveraging Conventional Neural Network 5.3 Role of Hyper-Spectral Data 5.3.1 Farm Based 5.3.2 Crop Based 5.3.3 Advanced HSI Processing 5.4 Potential Challenges and Strategies to Overcome the Challenges 5.5 Current and Future Scope 5.6 Conclusion References 6. An Ameliorated Methodology to Predict Diabetes Mellitus Using Random Forest 6.1 Motivation to Use the "R" Language to Predict Diabetes Mellitus? 6.2 Related Work 6.3 Collection of Datasets 6.3.1 Implementation Methods 6.3.1.1 Decision Tree 6.3.1.2 Random Forest 6.3.1.3 Naïve Bayesian Algorithm 6.3.1.4 Support Vector Machine (SVM) 6.4 Visualization 6.5 Correlation Matrix 6.6 Training and Testing the Data 6.7 Model Fitting 6.8 Experimental Analysis 6.9 Results and Analysis 6.10 Conclusion References 7. High Dimensionality Dataset Reduction Methodologies in Applied Machine Learning 7.1 Problems Faced with High Dimensionality Data: An Introduction 7.2 Dimensionality Reduction Algorithms with Visualizations 7.2.1 Feature Selection Using Covariance Matrix 7.2.1.1 Importing the Modules 7.2.1.2 The Boston Housing Dataset 7.2.1.3 Perform Basic Data Visualization 7.2.1.4 Pearson Coefficient Correlation Matrix 7.2.1.5 Detailed Correlation Matrix Analysis 7.2.1.6 3-Dimensional Data Visualization 7.2.1.7 Extracting the Features and Target 7.2.1.8 Feature Scaling 7.2.1.9 Create Training and Testing Datasets 7.2.1.10 Training and Evaluating Regression Model with Reduced Dataset 7.2.1.11 Limitations of the Correlation Matrix Analysis 7.2.2 t-Distributed Stochastic Neighbor Embedding (t-SNE) 7.2.2.1 The MNIST Handwritten Digits Dataset 7.2.2.2 Perform Exploratory Data Visualization 7.2.2.3 Random Sampling of the Large Dataset 7.2.2.4 T-Distributed Stochastic Neighboring Entities (t-SNE) - An Introduction 7.2.2.5 Probability and Mathematics behind t-SNE 7.2.2.6 Implementing and Visualizing t-SNE in 2-D 7.2.2.7 Implementing adn Visualizing t-SNE in 3-D 7.2.2.8 Applying k-Nearest Neighbors (k-NN) on the t-SNE MNIST Dataset 7.2.2.9 Data Preparation - Extracting the Features and Target 7.2.2.10 Create Training and Testing Dataset 7.2.2.11 Choosing the k-NN hyperparameter - k 7.2.2.12 Model Evaluation - Jaccard Index, F1 Score, Model Accuracy, and Confusion Matrix 7.2.2.13 Limitations of the t-SNE Algorithm 7.2.3 Principle Component Analysis (PCA) 7.2.3.1 The UCI Breast Cancer Dataset 7.2.3.2 Perform Basic Data Visualization 7.2.3.3 Create Training and Testing Dataset 7.2.3.4 Principal Component Analysis (PCA): An Introduction 7.2.3.5 Transposing the Data for Usage into Python 7.2.3.6 Standardization - Finding the Mean Vector 7.2.3.7 Computing the n-Dimensional Covariance Matrix 7.2.3.8 Calculating the Eigenvalues and Eigenvectors of the Covariance Matrix 7.2.3.9 Sorting the Eigenvalues and Corresponding Eigenvectors Obtained 7.2.3.10 Construct Feature Matrix - Choosing the k Eigenvectors with the Largest Eigenvalues 7.2.3.11 Data Transformation - Derivation of New Dataset by PCA - Reduced Number of Dimensions 7.2.3.12 PCA Using Scikit-Learn 7.2.3.13 Verification of Library and Stepwise PCA 7.2.3.14 PCA - Captured Variance and Data Lost 7.2.3.15 PCA Visualizations 7.2.3.16 Splitting the Data into Test and Train Sets 7.2.3.17 An Introduction to Classification Modeling with Support Vector Machines (SVM) 7.2.3.18 Types of SVM 7.2.3.19 Limitations of PCA 7.2.3.20 PCA vs. t-SNE Conclusion 8. Hybrid Cellular Automata Models for Discrete Dynamical Systems 8.1 Introduction 8.2 Basic Concepts 8.2.1 Cellular Automaton 8.3 Discussions on CA Evolutions 8.3.1 Relation between Local and Global Transition Function of a Spatially Hybrid CA 8.4 CA Modeling of Dynamical Systems 8.4.1 Spatially Hybrid CA Models 8.4.2 Temporally Hybrid CA Models 8.4.3 Spatially and Temporally Hybrid CA Models 8.5 Conclusion References 9. An Efficient Imputation Strategy Based on Adaptive Filter for Large Missing Value Datasets 9.1 Introduction 9.1.1 Motivation 9.2 Literature Survey 9.3 Proposed Algorithm 9.4 Experiment Procedure 9.4.1 Data Collection 9.4.2 Data Preprocessing 9.4.3 Classification 9.4.4 Evaluation 9.5 Experiment Results and Discussion 9.6 Conclusions and Future Work References 10. An Analysis of Derivative-Based Optimizers on Deep Neural Network Models 10.1 Introduction 10.2 Methodology 10.2.1 SGD 10.2.2 SGD with Momentum 10.2.3 RMSprop 10.2.4 Adagrad 10.2.5 Adadelta 10.2.6 Adam 10.2.7 AdaMax 10.2.8 NADAM 10.3 Result and Analysis 10.4 Conclusion References Section III: Applications of Data Science and Data Analytics 11. Wheat Rust Disease Detection Using Deep Learning 11.1 Introduction 11.2 Literature Review 11.3 Proposed Model 11.4 Experiment and Results 11.4.1 Dataset Preparation 11.4.2 Image Pre-processing 11.4.3 Image Segmentation 11.4.4 Discussion for the Model on Grayscale Images 11.4.5 Evaluating the Model on RGB Images 11.4.5 Result Comparison of the Model on RGB Images Based on Learning Rate 11.5 Conclusion References 12. A Novel Data Analytics and Machine Learning Model Towards Prediction and Classification of Chronic Obstructive Pulmonary Disease 12.1 Introduction 12.2 Literature Review 12.3 Research Methodology 12.3.1 Logistical Regression Model for Disease Classification 12.3.2 Random Forest (RF) for Disease Classification 12.3.3 SVM for Disease Classification 12.3.4 Decision Tree Analyses for Disease Classification 12.3.5 KNN Algorithm for Disease Classification 12.4 Experiment Results 12.5 Concluding Remarks and Future Scope 12.6 Declarations References 13. A Novel Multimodal Risk Disease Prediction of Coronavirus by Using Hierarchical LSTM Methods 13.1 Introduction 13.2 Related Works 13.3 About Multimodality 13.3.1 Risk Factors 13.4 Methodology 13.4.1 Naïve Bayes (NB) 13.4.2 RNN-Multimodal 13.4.3 LSTM Model 13.4.4 Support Vector Machine (SVM) 13.4.5 Performation Evaluation 13.4.5.1 Accuracy 13.4.5.2 Specificity 13.4.5.3 Sensitivity 13.4.5.4 Precision 13.4.5.5 F1-Score 13.5 Experimental Analysis 13.6 Discussion 13.7 Conclusion 13.8 Future Enhancement References 14. A Tier-based Educational Analytics Framework 14.1 Introduction 14.2 Related Works 14.3 The Three-Tiered Education Analysis Framework 14.3.1 Structured Data Analysis 14.3.1.1 Techniques for Structured Data Analysis 14.3.1.1.1 Correlation Analysis 14.3.1.1.2 Association Mining 14.3.1.1.3 Predictive Modeling 14.3.1.2 Challenges in Structured Data Analysis 14.3.2 Analysis of Semi-Structured Data and Text Analysis 14.3.2.1 Use Cases for Analysis of Semi-Structured and Text Content 14.3.2.2 Challenges of Semi-Structured/Textual Data Analysis 14.3.3 Analysis of Unstructured Data 14.3.3.1 Analysis of Unstructured Data: Study and Use Cases 14.3.3.2 Challenges in Unstructured and Multimodal Educational Data Analysis 14.4 Implementation of the Three-Tiered Framework 14.5 Scope and Boundaries of the Framework 14.6 Conclusion and Scope of Future Research Note References 15. Breast Invasive Ductal Carcinoma Classification Based on Deep Transfer Learning Models with Histopathology Images 15.1 Introduction 15.2 Background Study 15.2.1 Breast Cancer Detection Based on Machine Learning Approach 15.2.2 Breast Cancer Detection Based on Deep Convolutional Neural Network Approach 15.2.3 Breast Cancer Detection Based on Deep Transfer Learning Approach 15.3 Methodology 15.3.1 Data Acquisition 15.3.2 Data Preprocessing Stage 15.3.3 Transfer Learning Model 15.3.3.1 Visual Geometry Group Network (VGGNet) 15.3.3.2 Residual Neural Network (ResNet) 15.3.3.3 Dense Convolutional Networks (DenseNet) 15.4 Experimental Setup and Results 15.4.1 Performance Evaluation Metrics 15.4.2 Training Phase 15.4.3 Result Analysis 15.4.4 Comparison with Other State of Art Models 15.5 Discussion with Advantages and Future Work 15.5.1 Discussion 15.5.2 Advantages 15.5.3 Future Works 15.6 Conclusion References 16. Prediction of Acoustic Performance Using Machine Learning Techniques 16.1 Introduction 16.2 Materials and Methods 16.3 Proposed Methodology 16.3.1 Step 1: Data Preprocessing 16.3.2 Step 2: Fitting Regression Model 16.3.3 Building a Backward Elimination Model 16.3.4 Building the Model Using Forward Selection Model 16.3.5 Step 3: Optimizing the Regressor Model—Mean Squared Error 16.3.6 Step 4: Understanding the Results and Cross Validation 16.3.7 Step 5: Deployment and Optimization 16.3.7.1 Structural Parameters of Each Layer Material Is Shown in 16.4 Results and Discussions 16.4.1 Error Analysis and Validating Model Performance for All Test Samples 16.5 Conclusion References Section IV: Issue and Challenges in Data Science and Data Analytics 17. Feedforward Multi-Layer Perceptron Training by Hybridized Method between Genetic Algorithm and Artificial Bee Colony 17.1 Introduction 17.2 Nature-Inspired Metaheuristics 17.3 Genetic Algorithm Overview 17.4 Proposed Hybridized GA Metaheuristic 17.5 MLP Training by GGEABC 17.6 Simulation Setup and Results 17.7 Conclusion Acknowledgment References 18. Algorithmic Trading Using Trend Following Strategy: Evidence from Indian Information Technology Stocks 18.1 Introduction 18.2 Literature Survey 18.2.1 Data and Period of Study 18.3 Methodology 18.4 Results and Discussions 18.5 Conclusions 18.5.1 Future Scope References 19. A Novel Data Science Approach for Business and Decision Making for Prediction of Stock Market Movement Using Twitter Data and News Sentiments 19.1 Introduction 19.2 Review of Literature 19.3 Proposed Methodology 19.3.1 Sentiment Score 19.3.2 Labeling 19.3.3 Feature Matrix 19.3.4 Probabilistic Neural Network 19.4 Numerical Results and Discussion 19.4.1 Data Description 19.4.2 Statistical Measure 19.5 Simulation Results and Validation 19.5.1 Comparative Analysis over Existing and Proposed Decision-Making Methods 19.6 Conclusion and Future Enhancement References 20. Churn Prediction in the Banking Sector 20.1 Introduction 20.1.1 Problem Statement 20.1.2 Current Scenario 20.1.3 Motivation 20.1.4 Objective 20.2 Related Work 20.3 Methodology 20.3.1 Dataset 20.3.2 Proposed System for Customer Churn Prediction 20.4 Results 20.4.1 Analysis of Clustering of Churned Customers 20.5 Conclusion 20.6 Future Work References 21. Machine and Deep Learning Techniques for Internet of Things Based Cloud Systems 21.1 Introduction 21.1.1 Power of Remote Computing 21.1.2 Security and Privacy Policies 21.1.3 Integration of Data 21.1.4 For Hosting, Providers Remove Entry Barrier 21.1.5 Improves Business Continuity 21.1.6 Facilitates Inter-device Communication 21.1.7 Pairing with Edge Computing 21.1.8 How IoT and Cloud Complement Each Other? 21.1.9 Cloud and IoT: Which Is Better? 21.1.10 The Challenges Posed by the Cloud and IoT Together? 21.1.10.1 Handling an Outsized Amount of Knowledge 21.1.10.2 Networking and Communication Protocols 21.1.10.3 Sensor Networks 21.1.10.4 Security Challenges 21.2 Security Issues in IoT-Based Cloud Systems 21.2.1 Attacks in IoT 21.2.1.1 Active Attack 21.2.1.2 Passive Attack 21.3 Machine Learning and Deep Learning: A Solution to Cyber Security Challenges in IoT-Based Cloud Systems 21.3.1 Machine Learning and Deep Learning Techniques Introduction 21.3.1.1 A Tour of Machine Learning Algorithms 21.3.1.1.1 Regression Algorithms 21.3.1.1.2 Instance-Based Algorithms 21.3.1.1.3 Regularization Algorithms 21.3.1.1.4 Decision Tree Algorithms 21.3.1.1.5 Bayesian Algorithms 21.3.1.1.6 Clustering Algorithms 21.3.1.1.7 Association Rule Learning Algorithms 21.3.1.1.8 Artificial Neural Network Algorithms 21.3.1.1.9 Deep Learning Algorithms 21.3.1.1.10 Dimensionality Reduction Algorithms 21.3.1.1.11 Ensemble Algorithms 21.3.2 Machine Learning and Deep Learning Techniques Used in IoT Security 21.3.2.1 Supervised Machine Learning 21.3.2.1.1 Decision Trees 21.3.2.1.2 Support Vector Machines (SVMs) 21.3.2.1.3 Bayesian Theorem-Based Algorithms 21.3.2.1.4 K-Nearest Neighbor (KNN) 21.3.2.1.5 Random Forest (RF) 21.3.2.1.6 Association Rule (AR) Algorithms 21.3.2.1.7 Ensemble Learning (EL) 21.3.2.2 Unsupervised ML 21.3.2.2.1 K-Means Clustering 21.3.2.2.2 Principal Component Analysis (PCA) 21.3.2.3 Deep Learning (DL) Methods for IoT Security 21.3.2.3.1 Convolution Neural Networks (CNNs) 21.3.2.3.2 Recurrent Neural Networks (RNNs) 21.3.2.4 Unsupervised DL (Generative Learning) 21.3.2.4.1 Deep Auto Encoders (AEs) 21.3.2.4.2 Restricted Boltzmann Machines (RBMs) 21.3.2.4.3 Deep Belief Networks (DBNs) 21.3.2.5 Semi-Supervised or Hybrid DL 21.3.2.5.1 Generative Adversarial Networks (GANs) 21.3.2.5.2 Ensemble of DL Networks (EDLNs) 21.3.2.5.3 Deep Reinforcement Learning (DRL) 21.4 Conclusion References Section V: Future Research Opportunities towards Data Science and Data Analytics 22. Dialect Identification of the Bengali Language 22.1 Introduction 22.2 Previous Works 22.3 Proposed Methodology 22.3.1 Computation of Features 22.3.1.1 Feature Selection 22.3.1.1.1 Zero Crossing Rate (ZCR) Based Feature Computation 22.3.1.1.2 Mel Frequency Cepstral Coefficients (MFCCs) Based Feature Computation 22.3.1.1.3 Skewness-Based Feature Computation 22.3.1.1.4 Spectral Flux Based Feature Computation 22.3.2 Formation of Feature Vector and Classification 22.4 Experimental Results 22.4.1 Relative Analysis 22.5 Conclusion References 23. Real-Time Security Using Computer Vision 23.1 Introduction 23.1.1 Biometric 23.1.2 Computer Vision 23.1.3 Opencv Library 23.2 Data Security 23.3 Technology 23.3.1 Face Detection 23.3.2 Face Recognition 23.3.3 Haar Cascade Classifier 23.4 Algorithm 23.4.1 Algorithm to Capture the Image for Database 23.4.2 Algorithm to Recognize the Face 23.4.3 Algorithm to Train the Face Recognizer 23.4.4 Algorithm for Security 23.5 Result 23.6 Conclusion 23.7 Future Scope Reference 24. Data Analytics for Detecting DDoS Attacks in Network Traffic 24.1 Introduction 24.2 Background 24.3 Related Work 24.4 Methodology 24.4.1 Oversampling and Synthetic Sampling of Data 24.4.2 Detection of Stealthy DDoS attacks 24.4.3 Performance Evaluation by Ranking Machine Learning Algorithms 24.5 Result and Discussion 24.5.1 Datasets Used for Evaluation 24.5.2 Evaluation Metrics Used 24.5.3 Observations 24.6 Conclusion Notes References 25. Detection of Patterns in Attributed Graph Using Graph Mining 25.1 Introduction 25.2 Research Background 25.3 Literature Survey 25.4 General Definitions 25.4.1 Multi-relational Edge-attributed Graph 25.4.2 Multi-layer Edge-attributed Graph 25.4.3 Attributed Graph 25.5 Problem Definition 25.6 Proposed Approach 25.6.1 Pattern Length of 4, 5, and 25.6.1.1 For Length = 25.6.1.2 For Length = 25.6.1.3 For Length = 25.6.2 Node-Pair Generations 25.6.2.1 Node-Pair Generation for Three Attributed Line and Loop Patterns 25.6.2.2 Node-Pair Generation for Four Attributed Line and Loop Patterns 25.6.2.3 Node-Pair Generation for Four Attributed Star Patterns 25.6.2.4 Node-Pair Generation for Five Attributed Elongated Star Patterns 25.6.3 Pattern Detections 25.6.3.1 Three-Attributed Line Pattern 25.6.3.2 Three-Attributed Loop Pattern 25.6.3.3 Four-Attributed Line Pattern 25.6.3.4 Four-Attributed Loop Pattern 25.6.3.5 Four-Attributed Star Pattern 25.6.3.6 Five-Attributed Elongated Star Pattern 25.7 Proposed Algorithm for Detection of Patterns - Line, Loop, Star, and Elongated Star 25.7.1 Algorithm PDAGraph345( ) 25.7.2 Procedure for Node-Pair Assignment 25.7.3 Procedure to Create Three-Attributed Line and Loop Patterns 25.7.4 Procedure to Display Three-Attributed Line and Loop Patterns 25.7.5 Procedure to Create Four-Attributed Line and Loop Patterns 25.7.6 Procedure to Display Four-Attributed Line and Loop Patterns 25.7.7 Procedure to Create Four-Attributed Star Patterns 25.7.8 Procedure to Display Four-Attributed Star Patterns 25.7.9 Procedure to Create Five-Attributed Elongated Star Patterns 25.7.10 Procedure to Assign Node IDs of Five-Attributed Elongated Star Patterns 25.7.11 Procedure to Display Five-Attributed Elongated Star Patterns 25.7.12 Procedure to Generate Node-Pairs 25.7.13 Explanation of PDAGraph345( ) 25.8 Experimental Results 25.8.1 Using C++ Programming Language 25.8.1.1 Three-Attributed Line Pattern (1-2-3) 25.8.1.2 Three-Attributed Loop Pattern (2-3-4-2) 25.8.1.3 Four-Attributed Line Pattern (1-3-2-4) 25.8.1.4 Four-Attributed Loop Pattern (1-3-4-2-1) 25.8.1.5 Four Attributed Star Pattern (1-3-2-3-4) 25.8.1.6 Five-Attributed Elongated Star Pattern (1-2-3-4-3-2) 25.8.2 Using Python Programming Language 25.8.2.1 Three-Attributed Line Pattern (1-2-3) 25.8.2.2 Three-Attributed Loop Pattern (2-3-4-2) 25.8.2.3 Four-Attributed Line Pattern (1-3-2-4) 25.8.2.4 Four Attributed Loop Pattern (1-3-4-2-1) 25.8.2.5 Four-Attributed Star Pattern (1-3-2-3-4) 25.8.2.6 Five-Attributed Elongated Star Pattern (1-2-3-4-3-2) 25.9 Analysis of Experimental Results 25.10 Conclusion References 26. Analysis and Prediction of the Update of Mobile Android Version 26.1 Introduction 26.1.1 Mobile Fragmentation 26.1.2 Treble - Google 26.1.3 Security Fix Support and Android Update 26.2 Systematic Literature Survey 26.2.1 API Compatibility Issues and Android Updates 26.2.2 Android Updates and Software Aging 26.2.3 Android Updates and Google Play Store 26.2.4 Security Standards Hardware Rooted in Mobile Phones 26.2.5 Security Fixes and Android Update 26.2.6 Machine Learning and Android Antivirus Updates 26.2.7 Smells Detection in Android Using Machine Learning 26.2.8 Android Malicious Classification Using Various ML Algorithms 26.3 Existing Techniques 26.4 Methodology and Tools Used in Existing Techniques 26.5 Proposed System 26.5.1 Schematic Overview of Mobile Android Update Prediction and Analysis 26.5.2 Flow Chart Depicting Mobile Android Update Prediction and Analysis 26.5.3 Algorithm for the Prediction and Analysis 26.5.3.1 Algorithm for Linear Regression Model and R Programming 26.5.3.2 Algorithm for Logistic Regression Model 26.5.3.3 Algorithm for Decision Tree Model 26.5.4 Methodology 26.5.5 Software Packages Used 26.5.6 Dataset Description 26.5.6.1 Attribute and Values Information 26.5.6.2 Missing Attribute Values: None 26.6 Experimental Results and Discussions 26.6.1 Graphical Representation 26.7 Conclusions and Future Work References Appendix: Datasets Sample Attachments Index
Donate to keep this site alive
1. Disable the AdBlock plugin. Otherwise, you may not get any links.
2. Solve the CAPTCHA.
3. Click download link.
4. Lead to download server to download.