دانلود کتاب Advances in Data Science and Artificial Intelligence: ICDSAI 2022, IIT Patna, India, April 23 – 24 (Springer Proceedings in Mathematics & Statistics, 403)

فهرست مطالب :

Preface
Organization
Program Committee Chairs
Reviewers
Contents
Sky Detection in Outdoor Spaces
1 Introduction
2 Related Work
3 Algorithm
4 Dataset
5 Implementation Details
6 Result
7 Conclusion
8 Future Scope
References
Defining, Measuring, and Utilizing Student\'s Learning in a Course
1 Introduction
2 Related Work
3 Dataset Description and Preprocessing
4 Methodology
4.1 Calculating Learning Velocity
4.2 Classification Based on Learning Velocity
4.3 Guess Prediction
5 Results and Analysis
5.1 Learning Velocity Distribution
5.2 Guess Prediction Analysis
6 Conclusion
References
Holistic Features and Deep-Guided Depth-Induced Mutual-Attention-Based Complex Salient Object Detection
1 Introduction
2 Related Works
3 The Proposed Method
3.1 Non-Complementary Feature Aggregation
3.2 Depth-Induced Mutual Attention—DIMA
3.3 Complementary Features Fusion Model
Cross-Complementary Fusion (CF)
Intra-Complementary Features Aggregation (IFA)
3.4 Loss Function
4 Experiment and Result Analysis
4.1 Dataset and Evaluation Metrics
4.2 Implementation Details and Training Details
4.3 Comparison and Result Analysis
4.4 Ablation Analysis
Validation of Three-Stream Network
Effectiveness of Holistic Feature Space
5 Conclusion
References
Machine Learning Based Decision Support System for Resilient Supplier Selection
1 Introduction
2 Literature Review
3 Methodology
3.1 Modeling Purchase Orders Using Simulation
3.2 Synthetic Data Generation Using CTGAN
3.3 Data Preprocessing
3.4 Feature Engineering
3.5 ANN Model Training & Hyperparameter Tuning
4 Results and Discussion
4.1 Model for Predicting Probability of On-time Delivery
4.2 Model for Predicting Actual Date of Delivery
4.3 Discussion
5 Conclusion and Future Work
References
An Adaptive Task Offloading Framework for Mobile Edge Computing Environment: Towards Achieving Seamless Energy-Efficient Processing
1 Introduction
2 Motivation
3 Contributions
4 Related Work
5 Proposed Framework
5.1 Communication Layer
5.2 Decision Layer
5.3 Service Layer
5.4 Application Layer
6 Conclusion and Future Work
References
Road Surface Classification and Obstacle Detection for Visually Impaired People
1 Introduction
2 Methodology
2.1 Dataset and Preprocessing
2.2 Feature Extraction and Dimensionality Reduction
2.3 Classification
3 Results
4 Conclusion
References
A Survey on Semantic Segmentation Models for Underwater Images
1 Introduction
2 Semantic Segmentation Models
2.1 SegNet
2.2 PSP-Net
2.3 U-Net
2.4 DNN-VGG
2.5 DeepLabv3+
2.6 SUIM-Net
3 Comparative Analysis
4 Conclusion
References
An Interactive Dashboard for Intrusion Detection in Internetof Things
1 Introduction
2 Background
2.1 Intrusion Datasets
2.2 Related Work
3 Proposed Work
3.1 Layered Architecture
3.2 Preprocessing of Datasets
3.3 Feature Selection Using Voting Technique
3.4 Classification Algorithms
4 Experimental Setup and Results
4.1 Experimental Setup
4.2 Results
5 Conclusion
References
An Analogous Review of the Challenges and Endeavor in Suspense Story Generation Technique
1 Introduction
2 Review of Previous Work
2.1 Computerized and Suspense Story Generation
2.2 Motion Pictures and Imagery-Based Storytelling
2.3 Difficulties and Endeavors of Spinning the Story
3 Conclusion and Discussion
References
Friend Recommendation System Using Transfer Learning in the Autoencoder
1 Introduction
1.1 Autoencoder
Architecture of the Autoencoder
Parametrization of Autoencoder
Working of Autoencoder Network
Types of Autoencoder
2 Friend Recommendation
2.1 General Recommendation System
2.2 Different Models Used for Recommendation System
2.3 General Model for Recommendation System
2.4 Autoencoder for Recommendation System (Fig. 2)
2.5 Autoencoder Parameter for Recommendation System
2.6 Working of Autoencoder Model for Recommendation System
3 Transfer Learning
3.1 Pseudocode for Reference
4 Outcome of the Model
5 Conclusion and Future Work
References
Analysis on the Efficacy of ANN on Small Imbalanced Datasets
1 Introduction
2 Literature Review
3 Methodology
4 Dataset Analysis
5 Results
6 Conclusion
References
Lightweight and Homomorphic Security Protocols for IoT
1 Introduction
2 IoT Devices, Their Network, and Limitations
3 Need for a Lightweight Homomorphic Encryption
3.1 Need for a Lightweight Encryption
3.2 Need for a Homomorphic Encryption
4 Security Protocols for IoT Networks
4.1 Message Queuing Telemetry Transport (MQTT)
4.2 Constraint Application Protocol (CoAP)
4.3 IPv6 Over Low-Power Wireless Personal Area Networks (6LoWPAN)
4.4 ZigBee
5 A Study of Existing Lightweight and Homomorphic Cryptosystems
5.1 The ElGamal–Elliptic Curve Encryption Homomorphic Scheme
Key Generation
Encryption–Decryption
Homomorphic Nature
Security
5.2 Pure ElGamal Homomorphic Cryptosystem
Key Generation
Encryption–Decryption
Homomorphic Nature
Security
5.3 Integer-Based LHE for Mobile Cloud Networks
Key Generation
Encryption–Decryption
Homomorphic Nature
Security
5.4 Goldwasser–Micali Encryption Scheme
Key Generation
Encryption–Decryption
Homomorphic Nature
Security
5.5 Benaloh Cryptosystem
Key Generation
Encryption–Decryption
Homomorphic Nature
Security
5.6 Okamoto–Uchiyama Cryptosystem
Key Generation
Encryption–Decryption
Homomorphic Nature
Security
5.7 Unpadded RSA
Key Generation
Encryption–Decryption
Homomorphic Nature
Security
5.8 Paillier Cryptosystem
Key Generation
Encryption–Decryption
Homomorphic Nature
Security
5.9 Boneh–Goh–Nissim Encryption Scheme
Key Generation
Encryption–Decryption
Homomorphic Nature
Security
6 Conclusion
7 Future Work
References
Tool-Based Approach on Digital Vulnerability Management Hub (VMH) by Using TheHive Platform
1 Introduction
2 Related Work
3 Approach on Framework
3.1 Knowledge Collector
3.2 Asset Collector
3.3 Vulnerability Collector
3.4 Processing Module
4 Methodology
4.1 Integration with the Hive
Webhook Configuration of the Hive
ElastAlert Configuration
5 Simulations and Analysis
6 Conclusion and Future Works
References
Performance Analyzer for Blue Chip Companies
1 Introduction
2 Motivation
3 Literature Survey
4 Proposed Methodology
4.1 Data Collection and Filtering
4.2 Sentiment Analysis
4.3 Combining Results of Models
4.4 Stock Price Prediction
4.5 Pattern Detection and Analysis
5 Results and Discussion
5.1 Sentiment Analysis Result
5.2 Price Prediction Result
6 Conclusion
References
Strengthening Deep-Learning-Based Malware Detection Models Against Adversarial Attacks
1 Introduction
2 Background
3 Proposed Design
3.1 Creation of Malware Images
3.2 Autoencoder
3.3 GAN
3.4 Malware Classifier
4 Experiments
4.1 Dataset
4.2 Experiment Details
5 System Evaluation and Results
6 Conclusion and Future Scope
References
Video-Based Micro Expressions Recognition Using Deep Learning and Transfer Learning
1 Introduction
2 Literature Review
3 Methodology
3.1 Base Model
3.2 Preprocessing
3.3 Model Architecture
4 Results and Analysis
4.1 Accuracy
4.2 F1-Score
4.3 Area Under ROC Curve (AUC)
5 Conclusion
References
Trustworthiness of COVID-19 News and Guidelines
1 Introduction
2 Impact of Non-Trustworthy News and Guidelines
2.1 Impact of Fake News
2.2 Impact of False Guidelines
3 Measurement of Trustworthiness
3.1 Authenticated Source
3.2 Transparency
3.3 User Focus
3.4 Data Quality
4 Related Work
5 Dataset
6 Fake Detection Model Using Recurrent Neural Network (RNN)
6.1 Training Model
6.2 Performance Analysis
6.3 Hyperparameter Optimization
7 Implication and Future Work
7.1 Application of Solution
7.2 Future Research Direction
8 Conclusion
References
Detection of Moving Object Using Modified Fuzzy C-Means Clustering from the Complex and Non-stationary Background Scenes
1 Introduction
2 Related Works
3 Proposed Methodology
3.1 Clustering of Video Frames Using MFCM
Function Parameter
Computational Model
Fusion of Features Using MLOPM_CNN
Map Similarity Measure
4 Performance Examination
5 Conclusion
References
Deterrence Pointer for Distributed Denial-of-Service (DDoS) Attack by Utilizing Watchdog Timer and Hybrid Routing Protocol
1 Introduction
2 Related Works
3 Topology Distribution
3.1 Distance Vector Routing
3.2 Link-State Routing
3.3 Path Selection
4 Route Analytics
5 Energy Hoarding Algorithm
5.1 Implementing Energy Hoarding Algorithm Using OTcl
6 Methodology
7 Algorithm for Implementation
8 Simulations and Analysis
8.1 Analysis of the Evaluation Parameters
Throughput
Packet Delivery Ratio
Energy Consumption
Delay
9 Conclusion and Future Work
References
Modeling Logistic Regression and Neural Network for Stock Selection with BSE 500 – A Comparative Study
1 Introduction
2 Objectives
3 Data
3.1 Predictor Variables
3.2 Ranking Variable
3.3 Data Extraction
3.4 Datasets
3.5 Downsampling Training Samples
3.6 Response Variable
4 Cross Validation in Ratio Analysis: A Review
4.1 Set Aside Samples in Ratio Analysis
4.2 K-Fold Cross Validation in Ratio Analysis
5 Methodology
5.1 Experimental Design
Training Sets
6 Model Development
6.1 Model Configurations
6.2 Model Estimations
6.3 Model Applications
6.4 Model Performance Measures
6.5 Model Evaluations
7 Experimental Results
7.1 Model Validation
7.2 Model Optimization
7.3 Model Comparison
8 Conclusions
Appendix
References
Landslide Detection with Ensemble-of-Deep Learning Classifiers Trained with Optimal Features
1 Introduction
2 Literature Review
3 Problem Statement
4 Methodology
4.1 RNN
4.2 Bi-LSTM
4.3 Bi-GRU
5 Results & Discussion
6 Conclusion
References
A Survey Paper on Text Analytics Methods for Classifying Tweets
1 Introduction
2 Text Analytics Methods
3 Choosing Optimum Method
4 Exploring a Few Methods
4.1 Sentiment Analysis
4.2 Opinion Mining
4.3 Argument Mining
4.4 Stance Detection
5 Differentiating the Aforementioned Methods
6 Conclusion
References
A Survey on Threat Intelligence Techniques for Constructing, Detecting, and Reacting to Advanced Intrusion Campaigns
1 Introduction
2 Constructing Threat Intelligent Platform
2.1 Reasons for Threat Intelligence
2.2 Identifying Threat Actors
2.3 Tactics, Techniques, and Procedures
2.4 Indicators of Compromise
2.5 Threat Intelligent Sharing
2.6 Current Limitations and Challenges
3 Detection of Intrusion Attempts
3.1 IDS Technology Types Based on Their Detection Method
3.1.1 Signature-Based Intrusion Detection System (SIDS)
3.1.2 Anomaly-Based Intrusion Detection System (AIDS)
3.2 IDS Technology Types Based on Their Positioning Within the Computer System
3.2.1 Network Intrusion Detection Systems (NIDS)
3.2.2 Host Intrusion Detection Systems (HIDSs)
3.2.3 Protocol-Based Intrusion Detection Systems (PIDSs)
3.2.4 Hybrid Intrusion Detection Systems
3.3 Comparison of Model Performances on Public Datasets
4 Reacting to Intrusion Attempts
4.1 Comparison Features
4.1.1 Administrator\'s Involvement
4.1.2 Reaction Type
4.1.3 Scalability
4.1.4 Time Complexity
4.1.5 Response Policy
4.2 Survey of IRS in Literature
4.2.1 NICE: Network Intrusion Detection and Countermeasure Selection in Virtual Network Systems
4.2.2 IDAR: An Intrusion Detection and Adaptive Response Mechanism for MANETs
4.2.3 DOCS: Dynamic Optimal Countermeasure Selection for Intrusion Response System
4.2.4 ACIRS: A Risk Mitigation Approach for Autonomous Cloud Intrusion Response System
4.2.5 Deep Q-Learning: A Hybrid Model-Free Approach for the Near-Optimal Intrusion Response Control of Non-Stationary Systems
4.3 Comparison of Selected IRS from Literature
5 Conclusion
References
Generalizing a Secure Framework for Domain Transfer Network for Face Anti-spoofing
1 Introduction
2 Related Works
2.1 Anti-spoofing Methods
2.1.1 From RGB to Depth: Domain Transfer Network for Face Anti-spoofing 24:r1
2.1.2 Detection of Spoofing Medium Contours for Face Anti-spoofing 24:r2
2.1.3 Face Anti-spoofing via Stereo Matching 24:r3
2.1.4 Attention-Based Two-Stream Convolutional Networks for Face Spoofing Detection 24:r4
2.1.5 Attention-Based Two-Stream Convolutional Networks for Face Spoofing Detection 24:r5
2.2 Image Re-colourization
2.3 Image Encryption
3 Shortcomings of Existing Solutions
4 The Proposed Framework
4.1 Image Re-colourization 24:r7
4.2 Anti-spoofing Detection 24:r1
4.2.1 Domain Transfer
4.3 Image Encryption 24:r6
5 Comparisons with Other Methods
6 Conclusions and Future Work
References
Survey on Game Theory-Based Security Framework for IoT
1 Introduction
2 Threats and Security Requirements in IoT
2.1 Constraints of Hardware in IoT
2.2 Evaluation of Security Schemes
2.3 Threats in IoT
3 Game Theory in IoT
4 Game Theory-Based Security Models
4.1 Game Theory-Based Trust Model
4.1.1 Analysis of General Trust-Based Model
5 Conclusion
References
Survey: Intrusion Detection for IoT
1 Introduction
2 Internet of Things (IoT)
2.1 Security Challenges of IoT
2.1.1 Factors Affecting IoT Security
2.1.2 Emerging Security Issues
3 Intrusion Detection System (IDS)
3.1 Types of IDS
3.1.1 Network-Based IDS
3.1.2 Signature-Based IDS
3.1.3 Anomaly-Based IDS
3.1.4 Distributed-Based IDS
3.1.5 Host-Based IDS
4 IDS for IoT
4.1 Placement Strategies
4.1.1 Distributed IDS Placement
4.1.2 Centralised IDS Placement
4.1.3 Hybrid IDS Placement
4.2 Detection Methods
4.2.1 Signature-Based Approach
4.2.2 Anomaly-Based Approach
4.2.3 Specification-Based Approach
4.2.4 Hybrid Approach
5 Security Threats
5.1 Botnets
5.2 Denial of Service (DoS)
5.3 Man in the Middle (MITM)
5.4 Identity and Data Theft
5.5 Social Engineering
5.6 Routing Attacks
6 Analysis of Suitable IDS
7 Conclusion and Future Work
References
Human-in-the-Loop Control and Security for Intelligent Cyber-Physical Systems (CPSs) and IoT
1 Introduction
1.1 Internet of Things and Cyber-Physical Systems: A Common Ground
1.2 Introducing the Idea of Human-in-the-Loop with IoT/CPSs
2 Related Work
2.1 Security and Privacy Issues in IoT and CPS
2.2 Human-in-the-Loop-Based Security Models of CPS and IoT
3 A Security Framework to Protect Systems
3.1 Framework Components
3.1.1 Detection
3.1.2 Alarm
3.1.3 Analysis
3.1.4 Response
3.1.5 Communication
3.2 Comparison with Existing Approaches in HITLIoT/HITLCPS Security
4 Conclusion and Future Work
References
Survey: Neural Network Authentication and Tampering Detection
1 Introduction
2 Image Tampering: Overview
3 Neural Network Role in Authentication and Tampering
3.1 Digital Watermarking
3.1.1 Watermarking Techniques
3.1.2 Watermarking-Based Tampering Detection
3.1.3 Neural Network-Aided Image Watermarking
3.2 Image Authentication Based on Neural Networks
3.3 Image Tampering Detection Based on Neural Networks
3.3.1 Forgery Detection Based on DCT-CNN
3.3.2 Copy–Move Forgery Detection Based on SIFT-PCA: (Scale Invariant Feature Transform—Principal Component Analysis)
3.3.3 Forgery Detection Based on DWT (Discrete Wavelet Transforms)
4 Conclusion and Future Works
References
Misinformation Detection Through Authentication of Content Creators
1 Introduction
2 Related Work
3 Background
3.1 X.509 Certificates
3.2 Blockchain
3.3 Natural Language Processing
3.4 Image Matching Techniques
4 Proposed Model
4.1 Issue of Certificates
4.2 Authentication and Content Verification
4.3 Workflow of the Model
5 Comparison and Impact
6 Conclusion and Future Work
References
End-to-End Network Slicing for 5G and Beyond Communications
1 Introduction
2 Related Work
3 System Models
4 Results
5 Conclusion
References
Transparency in Content and Source Moderation
1 Introduction
2 Related Works
3 Methodology
3.1 Guidelines and Standard Protocol
3.2 Initial Moderation Through NLP Model
3.3 Weighted Voting Algorithm
3.4 Trust Calculation Algorithm
4 Analysis
4.1 Justification of NLP Model
4.2 Comparison with Other Moderation Systems
5 Conclusion and Future Work
References
A New Chaotic-Based Analysis of Data Encryption and Decryption
1 Introduction
2 Motivation
3 Literature Review
3.1 Advanced Encryption Standard (AES)
3.2 Rivest-Shamir-Adleman (RSA)
4 Methodology of CRSA
5 Analysis
6 Experimental Results
7 Conclusion
References
Trust and Identity Management in IOT
1 Introduction
2 Related Work
2.1 Identity Access Management (IdMS) in IoT
2.2 Trust Management in IoT
2.3 Identity Verification, Validation, and Secure Session Key Exchange in an IoT Cluster
3 Concepts in Identity Management
3.1 Characteristics of an IdMS
3.1.1 Security
3.1.2 Scalability
3.1.3 Interoperability
3.2 The Interdependence of Trust Management and IdMS
4 Trust Management Framework
4.1 Proposed Improvisation of the Above Method for Trust Calculation
5 Proposed Cryptographic Secure Session Key Exchange and Identity Verification Model
5.1 Architecture of the Model
5.2 Diffie–Hellman Algorithm for Secure Master Key (Km) Exchange
5.3 Sequential Flow of the Proposed Model
5.4 Resistance to Common Attacks
5.4.1 Man-in-the-Middle (MiTM) Attack
5.4.2 Relay Attack
5.4.3 Replay Attack
6 Conclusions and Future work
References
Plant Pest Detection: A Deep Learning Approach
1 Introduction
2 Literature Survey
3 Proposed System
3.1 Proposed System Architecture
3.2 Deep Learning Model
4 Implementation and Result Analysis
5 Conclusion and Future Scope
References
S.A.R.A (Smart AI Refrigerator Assistant)
1 Introduction
2 Literature Survey
3 Implementation
3.1 Data Mining
3.2 Preprocessing and Tokenization
3.3 Creation of Word Embeddings
3.4 Topic Modeling
3.5 BERT
3.6 Text Ranking
3.7 Querying Algorithm
4 Results
5 Conclusion
6 Future Scope
References
A Location-Based Cryptographic Suite for Underwater Acoustic Networks
1 Introduction
2 Related Work
3 Network Architecture
4 Proposed Algorithm for UANs
4.1 Communication Flow and Key Mapping
4.2 Round Key Generation
4.3 Encryption Algorithm
4.4 Decryption Algorithm
5 Comparative Study and Performance Analysis
6 Conclusion and Future Work
References
Index