Advances in Natural Computation, Fuzzy Systems and Knowledge Discovery: Proceedings of the ICNC-FSKD 2021 (Lecture Notes on Data Engineering and Communications Technologies, 89)

Preface
Organizing Committee
	General Chairs
	Program Chairs
	Organizing Chairs
	Publication Chairs
	Publicity Chairs
	Program Committee
Contents
Natural Computation: Theory and Algorithms
RRS: An Explainable Model Scale Search Strategy
	1 Introduction
	2 Model Scale Search Strategy
		2.1 Explainable Preprocessing Method
		2.2 Episodic Reinforced Sequence
	3 Experiment
		3.1 Comparative Verification Based on Interpretable Denoising Algorithms
		3.2 Correctness Verification Based on Episodic Enhanced Sequence
		3.3 Comparative Verification Analysis Based on Episodic Enhanced Sequence Method
		3.4 Conclusion
	References
Detection of Small Targets in Aerial Images Based on Feature Fusion
	1 Introduction
	2 MF-SSD Small Target Detection Network
		2.1 Small Target Detection Processing Model MF-SSD
		2.2 Design of Feature Fusion Sub-network
		2.3 The Receptive Field Enhancement Module
		2.4 Improved Feature Extraction Network
	3 Experiment and Analysis of Target Detection in Aerial Images
		3.1 Experimental Environment and Settings
		3.2 Experimental Results and Analysis
	4 Conclusion
	References
Intrusion Detection Based on LSTM and Random Forests
	1 Introduction
	2 Model Design
	3 Dataset
	4 Binary Classification Algorithm
		4.1 Activation Function
		4.2 Pooling Layer
		4.3 LSTM Layer
		4.4 Experimental Results
	5 Multiple Classification Algorithms
		5.1 Decision Tree
		5.2 Random Forests Model
		5.3 Experimental Results
	6 Conclusion
	References
Research and Hardware Implementation of Binocular Vision Obstacle Avoidance for UAV
	1 Introduction
	2 Principle of Binocular Stereo Vision
	3 Binocular Camera Calibration
	4 Canny Operator Edge Detection
	5 OpenCV-Based Binocular Vision Stereo Matching
		5.1 Stereo Matching Algorithm
		5.2 Stereo Calibration and Stereo Matching
	6 Conclusion
	References
Parameter Estimation of Equivalent Circuit Model for Lithium Batteries
	1 Introduction
	2 A Second-order RC Equivalent Circuit Model
	3 External Characteristics Experiment
		3.1 Different Multiplier Discharge Test
		3.2 Charging Experiment at Different Temperature
	4 Model System Estimation
		4.1 HPPC Discharge Test
		4.2 Parameter Estimation
		4.3 Model Building and Validation
	5 Conclusion
	References
Improved U-Net Network for Segmentation on Femur Images
	1 Introduction
	2 Improved U-Net Model
		2.1 Improved Residual Module
		2.2 Improved Jump Connection
		2.3 Improved Loss Function
	3 Experiment and Analysis
		3.1 Dataset
		3.2 Evaluation Indicators
		3.3 Network Structure Improvement Comparison Experiment
		3.4 Loss Function Comparison Experiment
		3.5 Comparison Experiments with Existing Methods
	4 Conclusion
	References
A 14-Bit 2MS/s Digital Self-calibrating SAR ADC Design and Simulation
	1 Introduction
	2 14Bit Self-calibrating SAR ADC Structure and Working Principle
		2.1 Overall Structure
		2.2 Calibrate DAC
		2.3 Working Principle
		2.4 Calibration Mode
	3 Simulation Results
	4 Conclusion
	References
A Windowed Interpolation Algorithm for High-Precision ADC Spectrum Testing
	1 Introduction
	2 The Principle and Design of Windowing Function
		2.1 Window Function Principle
		2.2 Design of Five-Term Fourth Order Hanning Window Functions
	3 Four Spectral Lines with Window Interpolation Algorithm
		3.1 A Windowed Interpolation Algorithm
		3.2 Analysis and Design of Four Spectral Line Interpolation Algorithm
	4 Simulation Experiment Analysis
		4.1 Test Basic Process
		4.2 Test Results and Analysis
	5 Conclusions
	References
Temperature Compensation Algorithms Based on Digital Infrared Thermopile Sensors Systems
	1 Introduction
	2 Infrared Temperature Sensors System Design
		2.1 Thermopile Temperature Sensors
		2.2 Design of Infrared Temperature Measurement Systems
	3 Experimental Platform Construction and Testing
	4 Algorithm Fitting and Analysis
		4.1 GA-BP Neural Network Algorithm
		4.2 Comparative Analysis of Experimental Data
	5 Conclusion
	References
Optimizing Deep-Learning Inference Operations
	1 Introduction
	2 Related Work
	3 Method
		3.1 Uniform Symmetric Quantizer
		3.2 Weight Sharing
	4 Experimental Results
		4.1 Datasets
		4.2 Evaluation Criteria
		4.3 Statistics on the Distribution of Weight Values in Different Networks
	5 Conclusion
	References
An Efficient MRI Impulse Noise Multi-stage Hybrid Filter Based on Cartesian Genetic Programming
	1 Introduction
	2 CGP Detection
		2.1 Feature Extraction Phase
		2.2 Cartesian Genetic Programming Model
	3 Noise Elimination Phase
		3.1 Adaptive Median Filter
		3.2 Shrinking Window Filter
		3.3 Edge-Preserving Filter
	4 Denoising Performance
	5 Conclusion
	References
Multi-objective Strategy Based on Rational Particle Swarm Optimization for Environmental/Economic Dispatch
	1 Introduction
	2 Brief Reviews
		2.1 Environmental Economic Dispatch Issues
		2.2 Objective Function
		2.3 Constraints
		2.4 Aggregate Objective Function
		2.5 Particle Swarm Algorithm
		2.6 Response Surface Method
	3 Improved Multi-objective Particle Swarm Algorithm
		3.1 The Second-Order Approximation Model
		3.2 Integrated Algorithm Model
		3.3 Algorithm Description
		3.4 Pareto Frontier Eclectic Solution Selection
	4 Algorithm Simulation and Analysis
		4.1 Analysis of Results
	5 Conclusions
	References
Express Service Composition Optimization Based on Improved Ant Colony Algorithm
	1 Introduction
	2 Proposed Problem
	3 Construction of Express Service Composition Models
		3.1 QoS Evaluation Model of Express Service
		3.2 The Objective Function and Constraints of Express Service Composition
	4 Improvement of Solution Algorithm
		4.1 Introduction of Ant Colony Algorithm
		4.2 Improvement of Ant Colony Algorithm
	5 Case Study and Its Result
		5.1 Experimental Design
		5.2 Data of the Experimental Example
		5.3 Experimental Results of Traditional Ant Colony Algorithm
		5.4 Experimental Results of Improved Ant Colony Algorithm
		5.5 Results Analysis
	6 Conclusion
	References
A Dynamic Multi-objective Evolutionary Algorithm Assisted by Kernel Ridge Regression
	1 Introduction
	2 Background Knowledge
		2.1 The Definition of the DMOPs
		2.2 Kernel Ridge Regression
	3 Proposed Algorithm
		3.1 Kernel Ridge Regression-Based Prediction Strategy
		3.2 MOEA/D-KRR Algorithm
	4 Computational Experiments
		4.1 Experimental Settings
		4.2 Experimental Results
	5 Conclusions
	References
Higher Order Preference on the Evolution of Cooperation on Barabási–Albert Scale-Free Network
	1 Introduction
	2 Model
		2.1 The Definition of Game Neighbor
		2.2 Prisoner's Dilemma Game Model Based on High-Order Preference
	3 Simulation Experiment Design and Results
		3.1 Simulation Experiment Design
		3.2 Simulation Experiment Results
	4 Analysis of Experimental Results
		4.1 Analysis of the Impact of High-Order Preference on the Cooperative Behavior of Different Connected Groups
		4.2 Analysis of the Relationship Between the Micro-Evolution and Macro-Performance of Individual Games
	5 Conclusion
	References
Nearest Neighbor Synthesis of CNOT Circuit Based on Matrix Transformation
	1 Introduction
	2 Background
	3 Feasible Strategies Analysis Based on Matrix Transformation
		3.1 NN Synthesis Based on the Boolean Matrix Flipped by the Center
		3.2 NN Synthesis Based on the Transpose Boolean Matrix
	4 Parallel NN Synthesis of CNOT Circuits
	5 Experiments and Result Analysis
	6 Conclusions
	References
State of Charge Estimation of Lithium Batteries Based on Extended Kalman Filter and Temperature Compensation
	1 Introduction
	2 Related Work
	3 Lithium Battery Model Establishment and Parameter Identification
		3.1 State-of-Charge Estimation Model
		3.2 Lithium Battery Equivalent Model
		3.3 Battery Model Identification Parameters
		3.4 Battery Model Identification Parameters
		3.5 EKF Algorithm
	4 Experimental Verification and Simulation Analysis
	5 Conclusion
	References
Improved GAN for Abnormal Flame Recognition Based on Siamese Network Structure
	1 Introduction
	2 Related Work
	3 Proposed Method
		3.1 Network Structure
		3.2 Loss Function
		3.3 Model Testing
	4 Experiments
		4.1 Dataset Preparation
		4.2 Results
	5 Conclusions
	References
Nine Electrodes Human Body Composition Model Based on SVM(SVM-BCM9)
	1 Introduction
	2 Data Acquisition System
	3 Raw Data Preprocessing
		3.1 Five Segment Model of Human Body
		3.2 BC Index
	4 SVM-BCM9
		4.1 SVM-BCM9 Algorithm Flow
	5 Experimental Results and Analysis
		5.1 Data Acquisition and Measurement
		5.2 Parameters Setting
		5.3 Comparison and Analysis of Human Experiment Results
	6 Conclusion
	References
A Self-adaptive Intrusion Detection Model Based on Bi-LSTM-CRF with Historical Access Logs
	1 Introduction
	2 Related Works
	3 Model and Training Process
		3.1 LSTM Layer
		3.2 Bi-LSTM Layer
		3.3 CRF Layer
	4 Experiment
		4.1 Dataset Preparation
		4.2 Process Description
		4.3 Results Discussion
	5 Conclusion
	References
A Left-Looking Sparse Cholesky Parallel Algorithm for Shared Memory Multiprocessors
	1 Introduction
	2 Background
		2.1 Sparse Column-Cholesky Algorithm
		2.2 Elimination Tree
		2.3 Supernode
		2.4 Description of the Supernodal Method
	3 Parallel Algorithm
	4 Thread Pool Mechanism
	5 Experiments and Results
		5.1 Test Environment
		5.2 Test Data
		5.3 Test Results
	6 Conclusion
	References
Improved Single Stage Bridge Crack Detection Algorithm
	1 Introduction
	2 Related Algorithms
		2.1 YOLOv3 Network Structure
		2.2 Anchor Box Clustering Based on K-means++
		2.3 Improvement of Network Structure
	3 Experiment and Analysis
	4 Conclusion
	References
Video Image Processing Based Intersection Vehicle Queue Length Detection
	1 Introduction
	2 Vehicle Queue Length Detection
		2.1 Vehicle Motion Detection
		2.2 Vehicle Presence Detection
	3 Calculate Vehicle Queue Length
	4 Experimental Analysis
	5 Conclusion
	References
Moving Vehicle Detection Combining Edge Detection and Gaussian Mixture Models
	1 Introduction
	2 Motion Vehicle Detection Algorithm in This Paper
		2.1 Video Image Pre-processing
		2.2 Vehicle Detection Algorithms
		2.3 Video Image Post-processing
	3 Experimental Results and Analysis
		3.1 Experimental Environment
		3.2 Performance Evaluation Indicators
		3.3 Experimental Results and Comparative Analysis of Moving Target Detection
	4 Conclusion
	References
An Adaptive Weight A* Algorithm Applied to UAV Path Planning
	1 Introduction
	2 Traditional A* Algorithm
	3 The Improved A* Algorithm
		3.1 Heuristic Function
		3.2 Extended Search Neighborhood
		3.3 Adaptive Weighting Strategy
		3.4 Rewiring
	4 Simulation Analysis
		4.1 Comparison Experiment
		4.2 Verification Experiments
	5 Conclusion
	References
Research on Extraction of Light Strip Centers Based on Legendre Moments
	1 Introduction
	2 Pretreatment
		2.1 Difference Image Methods
		2.2 Region Growing Algorithm
	3 Stripe Center Extraction
		3.1 Legendre Moments
		3.2 Cubic Hermite Splines
	4 Experimental Results and Analysis
	5 Conclusions
	References
Natural Computation Applications
Exploring Level-Wise Interpolationto Improve Lossy Compression Ratio for AMR Applications
	1 Introduction
	2 Motivation
	3 Level-Wise Interpolation
		3.1 Overview
		3.2 Interpolation Method Comparison
	4 Evaluation
		4.1 Experimental Setup
		4.2 Compression Ratios
	5 Conclusion
	References
Chinese Herbal Classification Based on Image Segmentation and Deep Learning Methods
	1 Introduction
	2 Material and Methods
	3 Results
	4 Discussion
	5 Conclusion
	References
A Fast and Efficient Vehicle Multi-target Tracking Algorithm for Edge Computing Devices
	1 Introduction
	2 Related Work
	3 Gain IoU Data Association and Feature Fusion Retracking Method
		3.1 Data Association Based on the Gain IOU Distance
		3.2 Re-tracking
	4 Experiments
	5 Conclusion
	References
A Sentiment Analysis Model for COVID-19 News in Chinese Social Networks
	1 Introduction
	2 Proposed Model
		2.1 System Architecture
		2.2 Scrapy Crawler
		2.3 Data Preprocessing
		2.4 Sentiment Analysis
	3 Experimental Evaluation
		3.1 Evaluation Criterion
		3.2 Two Baseline Methods
		3.3 Data Collection
		3.4 Comparison of Accuracy
		3.5 Sentiment Analysis for COVID-19 News
	4 Conclusions and Future Work
	References
A Time-Space Exchange Autoencoder for Image Super-Resolution
	1 Introduction
	2 Related Works
	3 System Architecture
		3.1 Network Architecture in This Paper
		3.2 Super-Resolution Process
	4 Experiment Result
		4.1 LR-HR Patch Pair Database
		4.2 SR Image Comparison Under Different Iterations
		4.3 Comparison of SR Image Achieved by Different Models
	5 Conclusion
	References
The Image Super-Resolution with Hopfield Neural Network
	1 Introduction
	2 Related Work
	3 Method
		3.1 Hopfield Neural Network with Multi-bitplane and Multi-connect Architecture
		3.2 Hopfield Network Based on BP Algorithm
	4 Experimental Results
		4.1 Datasets
		4.2 Comparisons with Super-Resolution Methods
	5 Conclusion
	References
Modified RK-EDA to Solve a Permutation-Based Spare Part Allocation Problem
	1 Introduction
	2 Problem Formulation
	3 Proposed Solution
		3.1 Targeted Site Indexing
		3.2 Improve the Way the Mean and Variance are Updated
	4 Experimental Results
	5 Conclusion and Future Work
	References
Futures Trend Forecasting Based on Attention Hybrid Neural Network
	1 Introduction
	2 Hybrid Neural Networks Based on Attention
		2.1 Convolutional Neural Network Design
		2.2 Attention-Based Bidirectional Long Short-Term Memory Network Design
		2.3 Hybrid Neural Network Design Based on Attention Mechanism
	3 Experiments
		3.1 Experimental Data
		3.2 Experimental Parameters and Evaluation Metrics
		3.3 Experimental Results and Analysis
	4 Conclusion
	References
Q-Learning Based DVFS for Multi-core Real-Time Systems
	1 Introduction
	2 System Model
		2.1 Power Model
		2.2 Task Characteristics
		2.3 Real-Time Task Model
	3 Q-Learning Based DVFS
		3.1 Q-Learning
		3.2 Architecture
	4 Experimental Results and Analysis
		4.1 Experimental Environment
		4.2 Performance on Variation of System Utilization
		4.3 Performance on Variation of Number of Tasks
	5 Conclusion
	References
Research and Design of Two-Wire Debugger Based on Reduced Instruction Set Computer
	1 Introduction
	2 Embedded Debugging Technology
		2.1 On-Chip Debugging Method
		2.2 Remote Debugger and Debugging Kernel
		2.3 JTAG Debug Interface
		2.4 Overall Structure of RISC-V Debugging Module
		2.5 Research on Two-Wire Debugging Protocol
		2.6 Two-Wire Debugging Protocol Write Operation Timing
		2.7 Two-Wire Debugging Protocol Read Operation Sequence
		2.8 Idle State of Two-Wire Debugging Protocol
		2.9 Two-Wire Debugging Protocol Error Response
	3 Design of Two-Wire Communication Protocol
		3.1 Header Module
		3.2 Receive Module
		3.3 Send Module
		3.4 Control Module
	4 Verification and Analysis
		4.1 Simulation Verification
		4.2 Functional Verification
		4.3 Logic Synthesis Result Analysis
	5 Conclusion
	References
Fuzzy Theory and Algorithms
Filters and Ideals in Q-lattices
	1 Introduction
	2 Preliminary
	3 Filters and Ideals in Q-lattices
	4 Weak Filters and Weak Ideals
	5 Homomorphisms
	References
A Limit Theorem on Fuzzy Set-Valued Random Variables in Fuzzy Metric Space
	1 Introduction
	2 Preliminaries and Notations
	3 Main Results
	4 Conclusion
	References
Interval Intuitionistic Hesitation Fuzzy Multi-attribute Decision Making Based on SRM-ELECTRE
	1 Introduction
	2 Correlation Theory of Interval Intuitive Hesitation Fuzzy Set Based on SRM and ELECTRE
		2.1 Interval Intuitionistic Hesitation Fuzzy Sets and Related Principles
		2.2 SRM-ELECTRE Evaluation Principle
	3 Multi-attribute Decision Method Based on SRM-ELECTRE
	4 Multi-attribute Decision Method Based on SRM-ELECTRE
		4.1 The Example Analysis
	5 Result Analysis
		5.1 Weight Method Analysis
		5.2 Analysis of Scheme Ranking Method
	6 Conclusion
	References
A Novel Unscented Particle Filter Based on Maximum Fuzzy Correntropy
	1 Introduction
	2 Maximum Fuzzy Correntropy Unscented Kalman Filter
		2.1 Maximum Fuzzy Correntropy Unscented Kalman Filter
	3 Maximum Fuzzy Correntropy Unscented Particle Filter
		3.1 Importance Sampling
		3.2 Residual Resampling
		3.3 Summary of Algorithm
	4 Simulation Results
		4.1 Experiment 1
		4.2 Experiment 2
	5 Conclusion
	References
A Multi-granularity Probabilistic Linguistic MABAC Method
	1 Introduction
	2 Preliminaries
	3 A Multi-granularity PLTS-MABAC Method
	4 Results Analysis
	5 Conclusions
	References
Normal Bipolar Fuzzy Subalgebras of BCK/BCI-Algebras
	1 Introduction
	2 Preliminaries
	3 Normal Bipolar Fuzzy Subalgebras
	4 Maximal and Completely Normal Bipolar Fuzzy Subalgebras
	5 Concluding Remarks
	References
Fuzzy Applications
A Fuzzy Testing Method for Gas-Related Vulnerability Detection in Smart Contracts
	1 Introduction
	2 Related Work
		2.1 Ethereum Virtual Machine
		2.2 Smart Contract
		2.3 Gas System
		2.4 Fuzzy Testing
		2.5 Three Gas-Related Vulnerabilities
	3 Vulnerability Detection Method
		3.1 Overview
		3.2 Feature Recognition
		3.3 Vulnerability Detection.
	4 Evaluation
		4.1 Experimental Platform
		4.2 Dataset
		4.3 Evaluating Indicator
	5 Results and Analysis
		5.1 Analysis of Target Detection Results
		5.2 Accuracy Comparison
	6 Conclusion
	References
Knowledge Discovery Foundations
Multi-granularity Knowledge Discovery Method Based on Layered Formal Context
	1 Introduction
	2 Preliminary
	3 Analysis of Multi-granularity Association Rules in the Layered Form Context
	4 Multi-granularity Association Rules Extraction in Layered Form Context
	5 Conclusion
	References
An Improved Canonical Correlation Analysis Method with Adaptive Graph Learning
	1 Introduction
	2 Related Works
		2.1 CCA
		2.2 Local Preserving CCA
	3 Proposed Method
	4 Experiments
	5 Conclusion
	References
Text Classification Based on ALBERT and Mutil-head Attention Capsule Network
	1 Introduction
	2 ABEA-ICapsule Text Classification Model
		2.1 ALBERT Layer
		2.2 BiGRU Layer
		2.3 Multi-head Self-attention
		2.4 Improved Capsule Network Model
		2.5 Group Normalization
		2.6 Dynamic Routing
	3 Experimental Results and Analysis
		3.1 Evaluation Index
		3.2 Experimental Parameters
		3.3 Comparative Experiment
	4 Conclusion and Prospect
	References
Classification Model Based Automatic Static Evaluation Method for C Programs
	1 Introduction
	2 Program Preprocessing
		2.1 Procedure Standardization
		2.2 Program Statement Dependency Graph
	3 Classification Model Based Evaluation Method
		3.1 Program Representation
		3.2 Hybrid Neural Network Classification Model Based on CST
	4 Experimental Results and Analysis
		4.1 Data Set
		4.2 Analysis and Evaluation of Experimental Results
	5 Conclusion
	References
Multi-label Text Classification Optimization Model Fusing ALBERT and LDA
	1 Introduction
	2 Related Work
	3 Our Model
		3.1 The Description of Multi-label Text Classification Model Diagram
		3.2 Input Presentation Layer
		3.3 Embedded Fusion Layer
		3.4 Feature Extraction Layer
		3.5 Feature Fusion Layer
		3.6 Output Layer
	4 DL-MLTC Model Time Complexity
		4.1 The Experimental Data
		4.2 The Experimental Setup
		4.3 The Evaluation Index
		4.4 The Experimental Results and Analysis
	5 Conclusion
	References
Definition and Segmentation of Key Clients in the N Metering Institution Based on Data Analysis
	1 Introduction
	2 Definition of Key Clients of Metering Institution
	3 Descriptive-Statistics-Based Segmentation of Key Clients in N Institution
		3.1 Key Client Segmentation According to Industries
		3.2 Key Client Segmentation According to Regions
	4 Segmentation of Key Clients in N Institution Based on K-means Clustering Algorithm
		4.1 Application Steps of K-means Clustering
		4.2 K-means Clustering Classification Calculation Process
		4.3 Analysis of K-means Clustering Classification Results
	5 Conclusion and Outlook
	References
Dual-Channel Microblog Emotion Analysis Based on RoBERTa-WWM and Multi-Head Attention
	1 Introduction
	2 DRWM Emotion Analysis Model
		2.1 Feature Presentation Layer
		2.2 Feature Extraction Layer
		2.3 Feature Fusion and Emotional Computing Layer
	3 Experimental Results and Analysis
		3.1 Experimental Data Set
		3.2 Evaluation Index and Parameter Setting
		3.3 Analysis of Results
	4 Conclusion and Propect
	References
A Network Combining Multi Dimension Scale and Feature Data for Automatic Melanoma Recognition from Skin Images
	1 Introduction
	2 Methods
		2.1 Data Enhancement and Preprocessing
		2.2 EfficientNet Network
		2.3 Characteristic Data Analysis
		2.4 Converged Network Efficient-Mix
		2.5 Loss Function Focal Loss
	3 Experiment
		3.1 Experimental Environment and Data Introduction
		3.2 Analysis of Experimental Results
	4 Conclusions
	References
An Energy Trading Privacy Protection Scheme Based on Blockchain and Non-interactive Zero-Knowledge Proof
	1 Introduction
	2 System Model
		2.1 Network Model
		2.2 Energy Trading Platform Based on Blockchain
		2.3 Privacy-Preserving Decentralized Energy Trading Scheme
	3 Performance Analysis
	4 Conclusion
	References
Incentive Strategies for Knowledge Sharing Between e-Commerce Supply Chain Enterprises from the Perspective of Trust
	1 Introduction
	2 Literature Review
		2.1 Influencing Factors of Knowledge Sharing Selected Under the Theory of Social Exchange
		2.2 Influencing Factors of Knowledge Sharing Selected Under the Theory of Social Capital
		2.3 Summary of Trust Theory Research
	3 Mechanism Analysis of e-commerce Supply Chain Knowledge Sharing Evolutionary Game
		3.1 Variables Definitions
		3.2 Model Variables and Assumptions
		3.3 Evolutionary Game Model
		3.4 The Dynamic Evolution Phase Diagram and Influencing Factors Analysis of the Game System
	4 Discussion of Incentive Strategies
	5 Conclusion
	References
Energy Trading System and Strategy Based on Blockchain and Deep Learning
	1 Introduction
	2 System Model
		2.1 Network Model
		2.2 Consensus Model
		2.3 Trading Mechanism
	3 Prediction Model
		3.1 Data Acquisition
		3.2 Prediction Model
	4 Experimental Results and Analysis
		4.1 System Performance
	5 Conclusion
	References
Random Weights Rough Neural Network for Glaucoma Diagnosis
	1 Introduction
	2 Methodology
		2.1 Dataset and Medical Features of Glaucoma
		2.2 Proposed Model
	3 Experimental Results
		3.1 Baselines and Experimental Settings
		3.2 Numerical Results
	4 Discussion
	5 Conclusion
	References
Emotion-Type-Based Global Attention Neural Network for Emotion-Cause Pair Extraction
	1 Introduction
	2 Related Work
	3 Methodology
		3.1 Task Definition
		3.2 Overall Description
		3.3 Emotion/Cause Clause Representation
		3.4 Emotion Type Prediction
		3.5 Emotion-Cause Pair Prediction
	4 Experiments
		4.1 Dataset and Implementation Details
		4.2 Compared Method
		4.3 Experimental Result
		4.4 Ablation Test
		4.5 Case Study
	5 Conclusions
	References
Face Data Discriminative Feature Extraction Based on Weighted Maximum Margin Criterion
	1 Introduction
	2 MMC
	3 WMMC
	4 Experimental Results and Analysis
	5 Conclusion
	References
Research on an Intelligent Decision System of Cold Chain Logistic for Fresh Agricultural Products
	1 Introduction
	2 Literature Review
	3 Construction of System
		3.1 Proposed Framework
		3.2 Data Collection
		3.3 Data Transmission
		3.4 Intelligent Decision Making
		3.5 User Application
	4 Subsystem Design
		4.1 Intelligent Order Management System
		4.2 Intelligent Warehouse Management
		4.3 Intelligent Transportation Decision System
		4.4 Intelligent Monitoring System
		4.5 Intelligent Delivery Cabinet
	5 Conclusion and Suggestions
		5.1 Conclusion
		5.2 Suggestions
	References
Facial Expression Recognition Based on Double-Channel Convolutional Neural Network
	1 Introduction
	2 Related Work
	3 Framework of the Proposed Method
		3.1 Data Pre-processing and Data Augmentation
		3.2 WMDCNN Network
		3.3 Weighted Mixture Double-Channel Output Network
	4 Experimental and Results
		4.1 Datasets
		4.2 Experimental Results of the WMDCNN Network
	5 Conclusions
	References
Weakly-Supervised Semantic Segmentation Based on Improved CAM
	1 Introduction
	2 CAM Method
	3 Improve the CAM
		3.1 Observation
		3.2 Significant Area Suppression
	4 Refinement of Labels
	5 Experiments
		5.1 The Experimental Setup
		5.2 Analysis
	6 Conclusion
	References
Prediction of Line Loss Rate Based on Improved BP Neural Network
	1 Introduction
	2 Factors Affecting Line Loss Rate of Power Grid
		2.1 Load Rate
		2.2 Power Supply Radius and Grid Structure
		2.3 Load Characteristic
	3 BP Prediction Algorithm and Its Implementation
		3.1 Neural Network Optimization
		3.2 Sample Set Data Processing
	4 Data Prediction and Analysis
	5 Conclusion
	References
Named Entity Recognition Based on Dependency Parsing and BiLSTM-CRF
	1 Introduction
	2 Related Work
		2.1 Word Vector and Character Vector
		2.2 BiLSTM
		2.3 Dependency Syntactic Analysis
	3 Approach
		3.1 Embedded Layer
		3.2 Feature Extraction Layer
		3.3 CRF Sequence Labeling
	4 Experimental
		4.1 Experimental Data and Evaluation Indicators
		4.2 Parameter Setting
		4.3 Analysis of Experimental Parameters
		4.4 Comparative Experiment and Analysis
	5 Conclusion
	References
Model Robust Optimization Method of Using Feature Pyramid
	1 Introduction
	2 A Robust Optimization Method for DNN Models
		2.1 Characteristic Pyramid Structure
	3 Experimental Results and Analysis
		3.1 The Experiment Design
		3.2 Experimental Procedure and Analysis
	4 Conclusion
	References
Fusing Multi-attention and Side Information for Distantly-Supervised Relation Extraction
	1 Introduction
	2 Related Work
	3 Methodology
		3.1 Task Definition
		3.2 Framework Overview
		3.3 Sentence Encoder
		3.4 Multiple-attention Mechanism and Side Information
		3.5 Model Training
	4 Experiments
		4.1 Dataset and Evaluation Metrics
		4.2 Model Setting
		4.3 Evaluation on Benchmark
		4.4 Ablation Results
	5 Conclusion
	References
An Augmented Reality Method Based on MobileNetV2SSD
	1 Introduction
	2 Related Work
		2.1 SSD Network
		2.2 MobileNetV2 Network
		2.3 SURF Algorithm
	3 Methodology
		3.1 Improved MobileNetV2SSD Target Detection Algorithm
		3.2 Transfer Learning
		3.3 The SURF Algorithm to Achieve Target Matching
		3.4 3-D Registration
	4 Model Training
		4.1 Experimental Platform
		4.2 Model Training
		4.3 The Evaluating Indicator
	5 Results and Analysis
		5.1 Analysis of Target Detection Results
		5.2 Analysis of Feature Matching Results
		5.3 Analysis of 3D Registration Results
	6 Conclusion
	References
An Augmented Reality Method Based on DeepSORT and Improved ORB
	1 Introduction
	2 Main Theory
		2.1 DeepSORT Algorithm
		2.2 ORB Algorithm
	3 Methodology
		3.1 Target Tracking
		3.2 Feature Extraction and Matching
		3.3 Three-Dimensional Registration
	4 Experimental Results and Analysis
		4.1 Experiment Platform
		4.2 Target Tracking Results and Analysis
		4.3 Feature Matching Results and Analysis
		4.4 Three-Dimensional Registration Results and Analysis
	5 Conclusion
	References
Fake News Detection Based on Dual Graph Attention Networks
	1 Introduction
	2 Related Works
		2.1 Fake News Detection
		2.2 Graph Neural Networks
	3 DGAT Model
		3.1 DGAT Model Overview
		3.2 Terminology Definitions
		3.3 Problem Definition
		3.4 Heterogeneous Information Embedding
		3.5 HIN Construction
		3.6 Information Aggregation
		3.7 Fake News Detection
		3.8 Model Training
	4 Experimental Results and Analysis
		4.1 Dataset
		4.2 Baseline Models
		4.3 Experimental Results
	5 Conclusion and Prospect
	References
Research on Marine Fish Recognition Algorithm Based on Op-CapsNet
	1 Introduction
	2 CapsNet
		2.1 The Basic Structure of CapsNet
		2.2 CapsNet Algorithm Structure
	3 Fish Recognition Model Based on Op-CapsNet
		3.1 Image Preprocessing
		3.2 Optimized Feature Extraction Structure
		3.3 Improved Partial Dynamic Routing Algorithm
		3.4 Fish Recognition Algorithm Based on Op-CapsNet
	4 Experimental Results and Analysis
		4.1 Experimental Data Sets and Preprocessing
		4.2 Experiment and Result Analysis
	5 Conclusion
	References
Adversarial Attacks Medical Diagnosis Model with Generative Adversarial Networks
	1 Introduction
	2 Adversarial Attacks Against Medical Model
		2.1 Adversarial Attacks Process
		2.2 Adversarial Attacks Architecture
	3 Construction Method of Generative Adversarial Networks
		3.1 Generator Network
		3.2 Discriminator Network
	4 Experimental Results
		4.1 Datasets
		4.2 Construction of Medical Diagnosis Models
		4.3 Adversarial Attacks Network Evaluation
	5 Conclusion
	References
A Behavior Recognition Algorithm Based on 3D-VGG-GAP Network
	1 Introduction
	2 Network Structure and Algorithm Design
		2.1 Network Design Process
		2.2 3D-VGG-GAP Network Structure
	3 Analysis of Network Parameters
	4 Experimental Results and Analysis
		4.1 Network Training Process and Test
		4.2 Test Indicators
	5 Conclusion
	References
Traffic Sign Recognition Based on Lenet-5 and Inception Modules
	1 Introduction
	2 Traffic Sign Recognition Model Based on Lenet-5 and Inception Module
		2.1 The Inception Structure
		2.2 Traffic Sign Recognition Model Based on Lenet-5 and Inception Module
	3 Experiment and Results Analysis
		3.1 Experimental Environment
		3.2 The Data Set
		3.3 Data Expansion
		3.4 Results and Analysis
	4 Conclusion
	References
Character Recognition Based on Composite Convolution Neural Networks and Coordinate Attention
	1 Introduction
	2 Related Work
	3 Character Recognition Based on Composite Convolution and Coordinate Attention
		3.1 Model Framework
		3.2 Composite Convolution
		3.3 Coordinate Attention
	4 Experiment and Results Analysis
		4.1 Data Sets
		4.2 Evaluation Methods
		4.3 Results Analysis
	5 Conclusion
	References
Research on Micro-expression Recognition Algorithm Based on Graph Convolutional Networks
	1 Introduction
	2 Micro-expression Recognition Model Based on Graph Convolutional Network
		2.1 Graph Convolutional Networks
		2.2 Design of Graph Convolutional Network Structure for Micro-expression Recognition
		2.3 Subset Division of Facial Muscle Nodes
	3 Experimental Design and Result Analysis
		3.1 Data Set Processing
		3.2 Network Structure and Parameter Configuration
		3.3 Experimental Results and Analysis
	4 Conclusion
	References
Sentiment Analysis of Barrage Text Based on ALBERT and Multi-channel Capsule Network
	1 Introduction
	2 Related Works
		2.1 ALBERT Layer
		2.2 BiGRU Layer
		2.3 Multi-channel Attention Mechanism
		2.4 Capsule Network
		2.5 Dynamic Routing
	3 Experimental Results and Analysis
		3.1 Evaluation Index
		3.2 Experimental Parameters
		3.3 Comparative Experiment
	4 Conclusion and Prospect
	References
An Improved CNN-LSTM Model Compression Pruning Algorithm
	1 Introduction
	2 Improved CNN-LSTM Pruning Algorithm
	3 Performance Test and Results of the Algorithm
		3.1 Data Set Introduction
		3.2 Experimental Settings
		3.3 Analysis of Experimental Results
	4 Conclusion
	References
An Aspect Term Extraction Method Based on BiLSTM-CRF with BERT Embedding
	1 Introduction
	2 Related Work
		2.1 Rule-Based Methods
		2.2 Machine Learning Methods
		2.3 Deep Learning Methods
	3 BiLSTM-CRF Aspect Term Extraction Model Based on BERT Embedding
		3.1 Model Framework
		3.2 BERT Embedding Layer
		3.3 BiLSTM Hidden Layer
		3.4 CRF Labeling Layer
	4 Experimental Analysis
		4.1 Dataset
		4.2 Parameter Setting
		4.3 Metrics
		4.4 Analysis of Experimental Results
	5 Conclusions
	References
Efficient Face Editing Generation Adversarial Networks Fused with Coordinate Attention
	1 Introduction
	2 Related Work
		2.1 Selective Transfer Units
		2.2 GhostNet
		2.3 Coordinate Attention Network
	3 Method
		3.1 Model Design Ideas
		3.2 Generator Network
		3.3 Discriminator Network
		3.4 Loss Function
	4 Experiment
		4.1 CelebA
		4.2 Experimental Parameter Settings
		4.3 Experimental Results and Analysis
	5 Conclusion
	References
Aspect-Level Sentiment Joint Detection Based on Graph Attention Network
	1 Introduction
	2 Related Work
	3 Proposed Method
		3.1 Word Embedding Layer
		3.2 BiGRU Layer
		3.3 Graph Attention Network Layer
		3.4 Output Layer
	4 Experiments
		4.1 Datasets
		4.2 Experimental Setup
		4.3 Compared Methods and Results Analysis
		4.4 Effectiveness Analysis of Each Module
	5 Summary and Outlook
	References
Low-Light Image Enhancement Combining U-Net and Self-attention Mechanism
	1 Introduction
	2 Related Work
		2.1 Self-attention Mechanism
		2.2 U-Net Network
	3 Method of this Article
		3.1 Model Design Ideas
		3.2 Generator Network Structure
		3.3 Discriminator Network Structure
		3.4 Construct Loss Function
	4 Experiments
		4.1 Dataset
		4.2 Experimental Details and Evaluation Index
		4.3 Experimental Results and Analysis
	5 Concluding
	References
Blockchain-Based Distributed Deep Learning Task Assignment Scheme
	1 Introduction
	2 Blockchain Network and Task Allocation Model
		2.1 System Model
		2.2 Optimization Objectives
	3 Task Allocation Algorithm Based on Reinforcement Learning
	4 Task Allocation System Design Based on Blockchain
		4.1 Task Allocation Management Structure
		4.2 Process of Resource Allocation Transaction
		4.3 Implementation Architecture
	5 Simulation Result and Discussion
	6 Conclusions
	References
AttnGAN++: Enhencing the Edge of Images on AttnGAN
	1 Introduction
	2 Data
	3 Methods
		3.1 The Structure of AttnGAN++
		3.2 Edge-Enhanced Deep Network
		3.3 Loss
	4 Experiments
		4.1 Quantitative Results
		4.2 Qualitative Results
	5 Conclusion
	References
Small Target Detection Method Based on a Multilayer Recursive Pyramid Network
	1 Introduction
	2 Related Work
		2.1 Single Shot Multi-box Detector Series
		2.2 Feature Pyramid Networks Series
		2.3 Feature Fusion
	3 MRPN Multilayer Recursive Pyramid Network
		3.1 Network Structure
		3.2 Selection of Anchor Size
		3.3 Loss Function
	4 Experiment and Results Analysis
		4.1 Experimental Environment
		4.2 The Model of Training Process
		4.3 Qualitative Analysis of Experimental Results
		4.4 Ablation Experiment
	5 Conclusion
	References
X-Ray Security Inspection Image Detection Based on a Multi-scale Feature Fusion Network
	1 Introduction
	2 SSD Target Detection Algorithm
	3 Target Detection Algorithm Based on Multi-scale Feature Fusion
		3.1 M-SSD Algorithm Structure
		3.2 Feature Fusion Module
		3.3 Asymmetric Convolution Module
	4 Experiments
		4.1 Data Set
		4.2 Experimental Environment
		4.3 Comparative Experiment
	5 Concluding Remarks
	References
COCOKG: A Multimodal Knowledge Graph Building on MSCOCO
	1 Introduction
	2 Related Work
	3 Building COCOKG
		3.1 MSCOCO Dataset
		3.2 Documents Summary
		3.3 Document Parsing
		3.4 Entity and Relation Extraction
	4 Enriching COCOKG
	5 Discussion
	6 Conclusion
	References
Deep Topological Reasoning with Region Connection Calculus
	1 Introduction
	2 Related Works
		2.1 Qualitative Spatial Reasoning and RCC Models
		2.2 Deep Learning Models for QSR
	3 Building the Benchmark Dataset
	4 Experiments
		4.1 Training from Scratch
		4.2 Transfer Learning
		4.3 Experiment Summary
	5 Conclusion
	References
Deep Learning Sentiment Analysis for MOOC Course Reviews
	1 Introduction
	2 Related Work
	3 Model Design
		3.1 Input Layer
		3.2 Pretreatment Layer
		3.3 Word Embedding Layer
		3.4 Bi-LSTM Layer
		3.5 Attention Layer
		3.6 CNN Layer
		3.7 Output Layer
	4 Experiment
		4.1 Data Set Acquisition and Processing
		4.2 Parameter Setting
		4.3 Experimental Comparison Model
		4.4 Analysis of Experimental Results
	5 Conclusion
	References
A Target Detection Algorithm in the Aerial Image Guided by the Density Map of the Target Center Point
	1 Introduction
	2 Algorithm Ideas
		2.1 Overall Idea of the Algorithm
		2.2 Density Map Generation Module
		2.3 Region Segmentation Module
		2.4 Target Detector Module
	3 Experiment and Analysis
		3.1 Datasets
		3.2 Quantitative Results
	References
Concept Relative Attention Based Deep Knowledge Tracing
	1 Introduction
	2 The Proposed Method
		2.1 Representation of Student Characteristics
		2.2 Loss Function
	3 Experiments
	4 Conclusion
	References
Two-Stage Multi-task Learning for Automatic Epilepsy Detection
	1 Introduction
	2 Materials and Methods
		2.1 Dataset
		2.2 Method
	3 Experiment
	4 Conclusion
	References
A Feature Embedding and Multi-dilation Convolutional Neural Network for Automatic COVID-19 Lesion Detection from Lung X-ray Images
	1 Introduction
	2 Data
		2.1 Gender Differences
	3 Methods
		3.1 Dilated Convolution
		3.2 COVID-Cascade RCNN
		3.3 Embedding
	4 Experiment
		4.1 Data Preprocessing
		4.2 Metrics
		4.3 Experimental Environment
		4.4 Experimental Results
	5 Conclusions
	References
A Detection Method for Pneumonia Lesions Based on Improved Feature Pyramid Network
	1 Introduction
	2 Detection Model of Pneumonia Lesions
	3 Improved Feature Pyramid Network
		3.1 First Stage Network of Improved FPN
		3.2 Second Stage Network of Improved FPN
	4 Experiment and Analysis
		4.1 Experiment Data
		4.2 Evaluation Criteria
		4.3 Experimental Setup
		4.4 Comparative Experiment
		4.5 Experimental Results and Analysis
	5 Conclusion
	References
Photo Mask Defect Detection Based on Generative Adversarial Network and Positive Samples
	1 Introduction
	2 Related Work
		2.1 Generative Adversarial Nets
		2.2 Image Repair Model Based on Positive Samples
		2.3 Autoencoder
	3 The Proposed Method
		3.1 Network Structure
		3.2 Objective
		3.3 Training Method
		3.4 Detecting Defects
	4 Experiment
		4.1 Data Set
		4.2 Experimental Platform and Parameter Setting
		4.3 Experimental Evaluation Index
		4.4 Results
	5 Conclusion
	References
A Low Illumination Images Enhancement Method Based on Conditional Generative Adversarial Networks with Self Attention Mechanism
	1 Introduction
	2 Generative Adversarial Networks
	3 Proposed Method
		3.1 Generator Networks Structure
		3.2 Self Attention Module
		3.3 Discriminator Improvement
	4 Experimental Analysis and Result Analysis
		4.1 Experimental Data
		4.2 Experimental Setting
		4.3 Subjective Evaluation
		4.4 Objective Evaluation
		4.5 Computational Efficiency
	5 Conclusion
	References
Grey Wolf-Particle Swarm Optimization with Back-Propagation Neural Network for Surface Duct Profile Estimation
	1 Introduction
	2 Calculation of Sea Clutter Power
	3 The GWO-PSO-BPNN Algorithm
		3.1 The Improved GWO
		3.2 GWO-PSO
		3.3 GWO-PSO-BPNN Algorithm
	4 Results Analysis
	5 Conclusion
	References
Knowledge Discovery in Specific Domains
Modified Newton Raphson Algorithm for Electrical Impedance Image Reconstruction
	1 Introduction
	2 Electrical Impedance Tomography Problem
		2.1 Basic Equation of Quasi-Static Electromagnetic Field
	3 Tikhonov Regularization Method
	4 Newton-Raphson Algorithm (NR)
		4.1 Newton Raphson Algorithm (NR) and Its Convergence Analysis
		4.2 Calculation of Gradient and Hessian Matrix
		4.3 Newton Raphson Algorithm Flow
	5 Modified Newton Raphson Algorithm (MNR)
		5.1 Gradient Descent Criterion
		5.2 Procedure of Modified Newton-Raphson Algorithm
	6 Experimental Results
		6.1 Simulation Experiment
		6.2 EIT Simulation Experiments
	7 Conclusions
	References
Mechanism of DNA Methylation in Stem Cells
	1 Summary of DNA Methylation
	2 The Role of DNA Methylation in Embryonic Stem Cells and Pluripotent Stem Cells
		2.1 Methylation in Embryonic Stem Cells and Reprogramming of iPSCs
		2.2 Role of TET Protein and 5hmC in Multipotent Stem Cell Reprogramming
	3 Role of DNA Methylation in Neural Stem Cells
		3.1 DNA Methylation in Adult Neural Stem Cells
		3.2 Role of DNA Hydroxymethylation and TET in Adult Neural Stem Cells
	4 Role of DNA Methylation in Germinal Stem Cells
		4.1 DNA Methylation and Demethylation of Adult Germ Stem Cells
		4.2 Role of TET and DNA Hydroxymethylation in Primordial Germ Cells
	5 Conclusions and Prospects
	References
Parallel Conditional Expectation Iteration Genomic Breeding Values Prediction Based on OpenMP
	1 Introduction
	2 Estimation of Breeding Value with CEI
		2.1 Statistical Model
		2.2 Prior Distribution of Marker Effect
		2.3 Prediction Using CEI
		2.4 Iteration Termination Criterion
	3 Parallel Genomic Prediction Based on OpenMP
	4 Experiment and Result Analysis
		4.1 Development Environment and Data
		4.2 Accuracy of Genomic Prediction
		4.3 Running Time
		4.4 Speedup Ratios in Parallel CEI
	5 Conclusion
	References
Visual Perception Law Under Bi-dimensional Optics Stimulation
	1 On the Law of Logarithm and the Law of Power Function of Sensation
	2 The Importance of Luminance
		2.1 The Importance of Luminance
		2.2 Incident Light and Reflected Light
	3 Quantification of Image Visual Quality Perception
		3.1 Sense Perception Laws of Human Visual Sense to Image Quality
		3.2 Visual Threshold and Blinding Threshold
		3.3 Boundary Conditions and Quantization Equation of Image Quality Perception
		3.4 Significance of Normalized Image Quality Sense
		3.5 Fuzzification on Visual Perception Quality of an Image
		3.6 Another Example of Quantifying Image Visual Perception Quality
	4 Conclusion
	5 Discussion
	Note Appendix
	References
Ear Recognition Based on Residual Network
	1 Introduction
		1.1 A Subsection Sample
	2 Ear Recognition Method Based on ResNet
		2.1 The Process of Ear Recognition
		2.2 Deep Residual Network
		2.3 Ear Recognition Method Based on ResNet18
	3 Experiment and Result Analysis
		3.1 USTB3 Human Ear Experiment Database and Experimental Environment
		3.2 Comparison of Ear Recognition Results with Different Parameters Based on ResNet18
		3.3 Comparison of Ear Recognition Methods Based on Different Depth Networks
		3.4 Performance Comparison Experiment with Classical Ear Recognition Method
	4 Conclusion
	References
Accelerating Group-By and Aggregation on Heterogeneous CPU-GPU Platforms
	1 Introduction
	2 Preliminary
		2.1 Coupled CPU-GPU Architectures
		2.2 Open Computing Language
	3 The Co-Processing Approach
		3.1 Observations
		3.2 The Co-processing Algorithm
	4 Performance Evaluation
	5 Related Work
	6 Conclusion
	References
HSMC: Hybrid Sentiment Method for Correlation to Analyze COVID-19 Tweets
	1 Introduction
	2 Methodology
		2.1 Proposed Method (HSMC)
		2.2 Data Crawling and Pre-processing
		2.3 Hybrid Sentiment Classification
		2.4 Correlation Detection
	3 Result Evaluation
		3.1 Data Description
		3.2 Hybrid Sentiment Evaluation
		3.3 Correlational Analysis
		3.4 Comparison with Other Studies
	4 Conclusion
	References
Self-supervised Pretraining for Covid-19 and Other Pneumonia Detection from Chest X-ray Images
	1 Introduction
	2 Background and Related Work
		2.1 Transfer Learning
		2.2 Self-supervised Contrastive Learning
		2.3 Object Detection
	3 Methodology
		3.1 Transfer Learning
		3.2 Contrastive Learning
	4 Experiments
		4.1 Dataset
		4.2 Experimental Setup
		4.3 Experimental Results
	5 Conclusion
	References
Research on Defect Classification and Analysis of Spacecraft Ground Software
	1 Introduction
	2 Defect Classification Method
		2.1 Common Defect Classification Methods
		2.2 MODC Classification
	3 Defect Attribute Analysis
		3.1 Domain Analysis
		3.2 Category Analysis
		3.3 Level Analysis
		3.4 Impact Analysis
		3.5 Priority Analysis
		3.6 Mode Analysis
		3.7 Introduction Analysis
		3.8 Detection Analysis
		3.9 Type Analysis
	4 Cause Analysis and Suggestions
		4.1 Poor Requirement Analysis and Specification
		4.2 Incomplete Software Design and Insufficient Feasibility Verification
		4.3 Basic Programming Rules are Commonly Violated in Coding
		4.4 Insufficient Software Testing and Large Differences Between Simulation Testing and Real Tasks
		4.5 Software Safety Analysis is Inadequate or Incomplete
	5 Conclusion
	References
MEMS Gyroscope Temperature Drift Compensation for SVR Optimized by Artificial Bee Colony
	1 Introduction
	2 ABC Optimization of SVR
		2.1 Support Vector Machine Regression (SVR)
		2.2 Artificial Bee Colony Algorithm (ABC)
		2.3 ABC Optimization of SVR MEMS Gyro Temperature Compensation Model
	3 Experimental Simulation and Result Analysis
	4 Conclusion
	References
Comprehensive Evaluation of High-Level Graduate School Based on Distanced Entropy-TOPSIS
	1 Inroduction
	2 Evaluation of Quality Evaluation of Graduate School Based on Distanced Entropy – TOPSIS
		2.1 Construction of Quality Evaluation Index System of Graduate School
		2.2 Evaluation Method Based on Distanced Entropy – TOPSIS
	3 An Example of Comprehensive Evaluation of High-level Graduate School based on Distanced Entropy – Topsis
	4 Conclusion
	References
Information Technology for Knowledge Discovery
Abnormal Telephone Recognition Based on Ensemble Learning
	1 Introduction
	2 Integrated Algorithm
	3 Experimental Analysis
		3.1 Data Source
		3.2 Data Preprocessing and Feature Engineering
		3.3 Analysis of Prediction Results
	4 Conclusion
	References
A Test System Design for JESD204B Transmitter Controller Chip
	1 Introduction
	2 Test System Design
		2.1 Test System Circuit PCB Design
	3 The Back-End Design of the Sender Controller
		3.1 Overview of the Functions of the Sender Controller
		3.2 Physical Layout Design of Transmitter Controller
	4 Hardware Circuit Design
	5 FPGA-based Chip Function Test
		5.1 Chip Test Plan Design
		5.2 Chip Test Method
		5.3 Experimental Results and Analysis
	6 Conclusion
	References
Text Difficulty Classification by Combining Machine Learning and Language Features
	1 Introduction
	2 Corpus
		2.1 Common Standards Corpus
		2.2 Corpus Extensions and Human Ratings
	3 Feature Extraction
		3.1 TFIDF-Based Feature Extraction
		3.2 Readability Formulas-Based Feature Extraction
		3.3 Transformer-Based Feature Extraction
	4 Experimental Evaluation
		4.1 Importance of Text Structure Features
		4.2 Comparison of Multiple Classifiers with Different Features
	5 Conclusion
	References
Multi-task Metric Learning for Text-independent Speaker Verification
	1 Introduction
	2 Related Work
		2.1 Baseline
		2.2 Classical Pair-Based Loss Functions
	3 Multi-task Metric Learning
		3.1 Data Selection and Loss Function
		3.2 Multi-task Learning Strategy
	4 Experiments and Discussion
		4.1 Experimental Settings
		4.2 Results
	5 Conclusion
	References
Combining Universal Speech Attributes into Deep Speaker Embedding Extraction for Speaker Verification
	1 Introduction
	2 Baseline X-Vector Architecture
		2.1 Multitask Learning with Universal Speech Attributes
		2.2 Generation of the NSA Units
		2.3 Multitask Learning with Cross-Stitch Units
	3 Experiments Setup
		3.1 Training Data and Evaluation Metric
		3.2 Input Features
		3.3 X-Vector Baseline System Configuration
		3.4 MTL with NSA Classification as an Auxiliary Task
	4 Result
		4.1 Comparison of Senones and NSA Units
		4.2 Comparison of Different Numbers of NSA Units
		4.3 Comparison of Different Numbers of Shared Layers
		4.4 Comparison Among Different MTL Frameworks
	5 Conclusion
	References
An Efficient Infrared Pedestrian Segmentation Algorithm Based on Weighted Maximum Entropy Thresholding
	1 Introduction
	2 Maximum Shannon Entropy Thresholding
	3 Weighted Entropy Thresholding Algorithm for Infrared Pedestrian Segmentation
	4 Experimental Results and Analysis
	5 Conclusions
	References
Topic Detection on Spoken Documents Using Deep-Feature of Acoustic Model
	1 Introduction
	2 TTD Based on ASR Transcripts
		2.1 ASR Module
		2.2 TTD Module
	3 TTD Based on Deep Feature
		3.1 Deep Feature Extractor
		3.2 DFHAN
	4 Experiments
		4.1 Conversational ASR and TTD System
		4.2 Topic Detection Based on DFHAN
	5 Conclusions
	References
Design of 4-bit ADC with Anti-interference for GNSS Receiver
	1 Introduction
	2 Sample-and-Hold Circuit
	3 Reference Voltage Generating Circuit
	4 Comparator Circuit
	5 Digital Coding Circuit
	6 Simulation
	7 Conclusion
	References
Energy-Aware Flow Table Strategy for Software Defined Wireless Power Communication Network Using Floyd Algorithm
	1 Introduction
	2 System Model and Problem Formulation
		2.1 System Model
		2.2 Problem Formulation
	3 Simulation
	4 Conclusions
	References
A Blockchain Privacy Protection Method Integrating Attribute Encryption and Group Signature
	1 Introduction
	2 Related Work
		2.1 Bilinear Mapping
		2.2 Verifiable Group Signature
		2.3 Attribute Encryption Strategy
	3 Blockchain Privacy Protection Scheme
		3.1 Storage Access Structure Design
		3.2 Authorization Center Design
		3.3 Blockchain Privacy Protection Scheme Model
		3.4 Program Description
	4 Analysis and Experimental Evaluation
		4.1 Safety Analysis
		4.2 Scheme Comparison
		4.3 Experimental Evaluation
	5 Conclusion
	References
Energy-Efficient Multi-domain Virtual Network Embedding Algorithm Based on Reinforcement Learning
	1 Introduction
	2 The System Model
		2.1 Multi-domain Heterogeneous Network Model
		2.2 Virtual Network Embedding Energy Consumption Model
	3 An Energy-Efficient Multi-domain Virtual Network Embedding Strategy Based on Reinforcement Learning
		3.1 Feature Extraction
		3.2 Reward Function
	4 Simulation and Performance Evaluation
		4.1 Experimental Environment Configuration
		4.2 Performance Evaluation Index
		4.3 Analysis of Simulation Results
	5 Conclusion
	References
A Review of Sparse Code Multiple Access Based on Deep Learning
	1 Introduction
	2 System Model
	3 Neural Network Implementation of SCMA
		3.1 Design of Encoder/decoder
		3.2 Codebook Design
	4 Summary and Prospect
	References
Neural Network Satellite Clock Bias Prediction Based on the Whale Optimization Algorithm
	1 Introduction
	2 The WOA Algorithm Principle
	3 Three Forecast Models
		3.1 Quadratic Polynomial Model
		3.2 BP Neural Network Model
		3.3 PSO-BP Neural Network Model
	4 WOA-BP Neural Network Model
	5 Experimental Result and Analysis
		5.1 Experiment One
		5.2 Experiment Two
	6 Concluding Remarks
	References
Cluster-TRnet: Jointed Model for Real-Time Traffic Identification with High Accuracy
	1 Introduction
	2 Related Works
	3 Algorithm Design
		3.1 Clustered Sub-model Data Processing
		3.2 Jointed CNN and LSTM Sub-model Data Processing
		3.3 Improved Jointed CNN and LSTM Model
	4 Experimental Design
		4.1 Experimental Data
		4.2 Experimental Evaluation Indicators
		4.3 Experimental Results and Analysis
	5 Conclusions
	References
Design and Implementation of a Lightweight Street View Segmentation Algorithm
	1 Introduction
	2 Related Work
		2.1 Residual Network
		2.2 Dilated Convolution
		2.3 U-Net
	3 Our Work
	4 Experiment
		4.1 Dataset
		4.2 Parameter Setting
		4.3 Results and Analysis
	References
Design and Implementation of the LAPS Protocol Mapper in the Military Ethernet Circuit
	1 Introduction
	2 System Design Scheme
	3 LAPS Mapping Circuit
	4 LAPS Demapping Circuit
	5 Data Transmission Efficiency Analysis
	6 Simulation Analysis
	7 Conclusion
	References
An Ethernet Mapping High-Level Data Link Control Circuit Design
	1 Introduction
	2 The Mapping Relationship Between Ethernet Frames and HDLC Frames
	3 HDLC Frame Sending Module
		3.1 Add Flag Bit Module
		3.2 CRC32 Generation Module
		3.3 Insert BER Module
		3.4 BitFill Module
		3.5 Scrambler Module
	4 HDLC Frame Receiver Module
		4.1 Descramble Module
		4.2 Frame Boundary Positioning Module
		4.3 BitDelete Module
		4.4 CRC32 Check Module
		4.5 BER Detection Module
		4.6 Delete Flag Bit Module
	5 Circuit Simulation
	6 Concluding Remarks
	References
Federated Learning-Based Codebook Design for Massive MIMO Communication System
	1 Introduction
	2 System Model and Problem Description
		2.1 System Model
		2.2 Problem Description
	3 FL for Codebook Design
		3.1 Data Collection
		3.2 Deep Network Architecture
	4 Simulation Results
	5 Conclusion
	References
Performance Analysis of MISO-NOMA Systems with Different Antenna Selection Schemes
	1 Introduction
	2 System Model
	3 Performance Analysis
		3.1 Outage Probability of Scheme I
		3.2 Outage Probability of Scheme II
		3.3 Asymptotic Analysis
		3.4 Throughput Performance
	4 Numerical Results and Discussions
	5 Conclusion
	References
An Dynamic Population Wolf Heuristic Algorithm Based User Service Subarray Mapping for Extra-Large Scale MIMO
	1 Introduction
	2 System Model
		2.1 System Model and Channel Model
		2.2 Problem Description
	3 DPWH Based Multi-user Service Subarray Mapping
		3.1 An Overview of WCA
		3.2 User Service Subarray Mapping Scheme for DPWH Algorithm
	4 Simulation Results
	5 Conclusion
	References
Millimeter Wave Beam Prediction Based on Compressed Sensing Features at Sub-6GHz
	1 Introduction
	2 Methodology
		2.1 The Model Architecture
		2.2 CS-Based Sub-6GHz Features Acquisition
		2.3 MmWave Beam Prediction Based on Sub-6GHz Features
	3 Deep Neural Network Model
	4 Simulations
		4.1 Dataset Acquisition
		4.2 Neural Network Training Parameters
		4.3 Performance Index
		4.4 Comparison Results
	5 Conclusions
	References
Blockage-Aware Backup Point Selection for Reliable mmWave Access via Markov Decision Processes
	1 Introduction
	2 System Model
		2.1 Network Model
		2.2 Backup Access Point for Blockage Robustness
		2.3 Markov Decision Process Definition
	3 Dynamic Backup Point Selection Algorithm
	4 Simulation Results
	5 Conclusion
	References
XGBoost Dynamic Detection for Data Race in Multithreaded Programs
	1 Introduction
	2 XGRace Data Race Instruction-Level Dynamic Detection Method
		2.1 Sim-hb Relationship and Lockset Algorithm Description
		2.2 XGBoost Data Race Detection Model
	3 Experimental Analysis
		3.1 Experimental Environment and Test Data Sets
		3.2 Experimental Results
	4 Conclusion
	References
Design and Realization of Automatic Test System for Fiber Optic Gyroscope
	1 Introduction
	2 Overall Scheme
		2.1 Functional Design
		2.2 System Hardware
		2.3 Software Design
	3 Related Technologies
		3.1 Data Collection and Storage
		3.2 Log Generation
		3.3 Multithreading Technology
	4 Technical Indicators
	5 Test Steps and Results
	6 Conclusion
	References
The Second-Order Equivalent Approximation of PID Control System Based on Dominant Pole Placement
	1 Introduction
	2 Problem Statement
	3 Root Locus Method
	4 Zero Position Analysis
	5 Example
	6 Conclusion
	References
Performance Optimization for Parallel SpMV on a NUMA Architecture
	1 Introduction
	2 Related Work
		2.1 Optimization of SpMV
		2.2 NUMA Architecture
	3 Sparse Matrix Partition Strategy
		3.1 Aligned Compressed Sparse Row
		3.2 Partition Strategy
		3.3 Implemention
	4 Experimentals
		4.1 Experiment Platform
		4.2 Experiment Data
		4.3 Experiment Results
	5 Conclusion
	References
Structure Optimization and Reconfigurable Design of CycleGAN
	1 Introduction
	2 Model Optimization
	3 Dynamic Reconfigure Design of S-CycleGAN
		3.1 The Dynamic Reconfigurable Scheme of S-CycleGAN
	4 Experimental Results and Performance Analysis
		4.1 Effect of Improved Model
		4.2 Performance of Reconfigurable
	5 Conclusion
	References
Hardware Algorithm Design of Deferred Shading
	1 Introduction
	2 Algorithm
		2.1 Z-value Interpolation Algorithm Based on Gravity Coordinate System
		2.2 Scanning Algorithm Based on Edge Function
		2.3 Perspective Correction Attribute Interpolation Algorithm Based on Gravity Coordinate System
	3 Hardware Design
		3.1 Vertex Sorting Module
		3.2 Parallel Scan Module
		3.3 Effect Demonstration
	4 Conclusions
	References
Shader Datapath Design of OpenGL-ES Lite
	1 Introduction
		1.1 Introduction to GPU and Shader
		1.2 Introduction to OpenGL-ES
	2 Overall Architecture of Datapath
		2.1 The Overall Structure
		2.2 Instruction Set
	3 Sub-Module Design
		3.1 Fixed Point Unit Design
		3.2 Floating Point Unit Design
	4 Performance Analysis
	References
Multi-Level Network-on-Chip Design Based on Embedded Unified Rendering GPU
	1 Introduction
	2 Network on Chip in Embedded GPU
	3 On-Chip Network Layered Design
	4 Determine the Number of Virtual Channels
		4.1 Time Distribution
		4.2 Spatial Distribution
		4.3 Injection Rate
	5 Experimental Results and Analysis
	6 Conclusion
	References
Research on the Storage and Sharing Model of Electronic License Based on Blockchain
	1 Introduction
	2 Related Work
		2.1 Blockchain
		2.2 El Gamal Algorithm
		2.3 PBFT Consensus Algorithm
	3 System Model
	4 Implementation Scheme Design
		4.1 License Data Structure
		4.2 Improved El Gamal Digital Signature Scheme
		4.3 Improved PBFT Consensus Algorithm
	5 Experimental Analysis
		5.1 Safety Analysis
		5.2 Performance Analysis
	6 Conclusions
	References
Chinese Named Entity Recognition Based on BERT and Neural Network
	1 Introduction
	2 Model Design
		2.1 Input Layer
		2.2 Convolution Layer
		2.3 RBiGRU
		2.4 Output Layer
	3 Experiment
		3.1 Experiment Setting
		3.2 Influence of Vector Dimension of BERT Character on Model
		3.3 The Influence of Depth of Residual Network on the Model
		3.4 Model Comparison
	4 Conclusion
	References
Gap Detection of Aircraft Bell-Shaped Tube Based on Convolutional Neural Networks
	1 Introduction
	2 Related Theory Knowledge
		2.1 Related Research
		2.2 YOLOv4 Algorithm Principle
	3 Improved YOLOv4 Algorithm
		3.1 Improvement of Anchor Box Size
		3.2 Improved YOLOv4 Network
	4 Experiments
		4.1 Experimental Dataset
		4.2 Experimental Conditions
		4.3 Experiments and Result Analysis
	5 Conclusions
	References
Design and Application of an Irregular Layout Report Tool
	1 Introduction
	2 Overall Design of Irregular Layout Report Tool
	3 Characteristics of Report Tool
	4 Process Design
		4.1 Analyze and Build the Presentation Data Source of the Report
		4.2 Analyze and Build the Presentation Data Source of the Report
		4.3 Establish the Relationship Between Data Source and Report Presentation
		4.4 Publish Reports and Store Report Data
	5 Application of Reporting Tools
	6 Conclusion
	References
Development, Design and Application of Intelligent Report
	1 Overview
	2 Ideas of Report Tool Design
	3 Integration of Report Tool
	4 Advantages of Report Tool Products
	5 Report Tool Architecture
	6 Features of the Reporting Tool
		6.1 Universal Display
		6.2 Indicator Definition
		6.3 Irregular Layout
		6.4 Detailed Summary
		6.5 Statistical Analysis
		6.6 Chart Display
		6.7 Report Drill Function
		6.8 Report Reuse
		6.9 Report Analysis
		6.10 Report Import and Export
	7 Summary of System Features
	8 Application Scenario Analysis
	References
A New Blockchain Consensus Algorithm for Virtual Power Plants
	1 Introduction
	2 System Model
		2.1 Virtual Power Plant
		2.2 Function Module
	3 Data Collection and Encryption
		3.1 Paillier Encryption of the Collected Data
		3.2 Transaction Verification
	4 Optimization of Blockchain Consensus Algorithm
		4.1 The Node's Weight
		4.2 Reward and Punishment Scheme of Weight
		4.3 Main Node Selection Process Based on Weight-Based Rewards and Penalties
		4.4 Divide Consensus Domain and Storage Domain Based on Weight
	5 Verification Analysis
		5.1 Experimental Environment
	6 Conclusions
	References
LLC Access Behavior Prediction Method for Multi-core Processors
	1 Introduction
	2 Cache Modeling Based on Stack Distance
		2.1 Stack Distance
		2.2 StatCC Model
	3 Experimental Results and Analysis
		3.1 Experimental Environment
		3.2 LLC Miss Rate Forecast
		3.3 Model Accuracy Analysis
	4 Conclusion
	References
A Novel Approach for Automatic Measurement of the Distal Femur Based on 3D Slicer Software
	1 Introduction
	2 Materials and Methods
		2.1 Statistical Model Construction of Femur
		2.2 Automatic Measurement System Based on 3D Slicer
	3 Experiments and Results
		3.1 Evaluating the Model Matching Method
		3.2 Evaluating the Matching Accuracy of Marker Points
		3.3 Evaluating Automatic Measurement Results
	4 Discussion
	5 Conclusion
	References
Container Network Performance Anomaly Detection Based on Extended Berkeley Packet Filter and Machine Learning
	1 Introduction
	2 Related Work
		2.1 Network Performance Monitoring and Data Collection
		2.2 Network Performance Anomaly Detection
	3 Design of Network Performance Anomaly Detection Model
		3.1 Overall Design
		3.2 Collect Network Performance Data Based on eBPF
		3.3 Network Performance Anomaly Detection Based on Machine Learning
	4 Experiment Analysis
		4.1 Lab Environment
		4.2 Experimental Steps
		4.3 Experimental Evaluation
		4.4 Experimental Results
	5 Conclusion
	References
SYN Flood Attack Detection and Defense Method Based on Extended Berkeley Packet Filter
	1 Introduction
	2 Attack Detection and Defense System Model
	3 Data Extraction
		3.1 SYN Flood Attack Analysis
		3.2 Network Data Extraction
	4 Detection Algorithm and Defense Measures
		4.1 Detection Algorithm
		4.2 Defense Measures
	5 Experimental Results and Analysis
		5.1 Experimental Environment
		5.2 Verification and Analysis of Experimental Results
	6 Conclusion
	References
Prediction of Low-Frequency Electromagnetic Environment of Subway
	1 Introduction
	2 The Establishment of Three-Dimensional Models of Vehicle Body
	3 Low-Frequency Magnetic Field Distribution Under Normal Operation Conditions
	4 Various Factors Affecting the Low-Frequency Magnetic Field Inside the Vehicle Body
		4.1 Influence of Vehicle Body Material on the Low-Frequency Magnetic Field Inside the Carriage
		4.2 Influence of High Voltage Wire Sheath on the Low Frequency Magnetic Field Inside the Vehicle Body
		4.3 Influence of the Magnitude of the Traction Current on the Low Frequency Magnetic Field Inside the Vehicle Body
	5 Conclusion
	References
Author Index