Intelligent Computing and Block Chain First BenchCouncil International Federated Conferences, FICC 2020, Qingdao, China, October 30 {u2013} November 3, 2020, Revised Selected Papers

This book constitutes the refereed post-conference proceedings of the Second BenchCouncil International Federated Intelligent Computing and Block Chain Conferences, FICC 2020, held in Qingdao, China, in October/ November 2020. The 32 full papers and 6 short papers presented were carefully reviewed and selected from 103 submissions. The papers of this volume are organized in topical sections on AI and medical technology; AI and big data; AI and block chain; AI and education technology; and AI and financial technology. Preface Organization Contents AI and Medical Technology BLU-GAN: Bi-directional ConvLSTM U-Net with Generative Adversarial Training for Retinal Vessel Segmentation 1 Introduction 2 Method 2.1 Image Preprocessing 2.2 The Proposed Architecture 3 Experiments 3.1 Datasets 3.2 Implementation Details 3.3 Results 4 Conclusion References Choroidal Neovascularization Segmentation Based on 3D CNN with Cross Convolution Module 1 Introduction 2 Method 3 Experiment 3.1 Dataset 3.2 Implementation Details 3.3 Evaluation Criterion 3.4 Result 4 Conclusion References Task-Free Recovery and Spatial Characterization of a Globally Synchronized Network from Resting-State EEG 1 Introduction 2 Methods 2.1 Experimental Procedures 2.2 Blind Source Separation of Resting-State EEG Data 2.3 Identification of the SOBI-Recovered gRSN Component 2.4 Hypothesis-Driven Source Modeling 2.5 Scalp Projection of the gRSN as Input to BESA 2.6 Iterative ECD Model Fitting Procedure 2.7 From ECD Coordinates to Anatomical Structures 2.8 Quantitative Characterization of the gRSN’s Spatial Configuration 2.9 Hits Vector-Based Visualization of Individual Differences 2.10 Statistical Analysis 2.11 The “Inverse Problem” 3 Results 3.1 Reliable Recovery of gRSN Components from Resting-State EEG 3.2 Variable Neural Generators Underlying the SOBI Recovered gRSN Component 3.3 Quantifying Cross-individual Variability in Network Configuration 3.4 Quantifying Within-Individual Variations in Network Configuration 4 Conclusion 4.1 gRSN: A Spatially Defined High-Dimensional Neural Marker 4.2 Individual Differences in Spatial Configuration of gRSN 4.3 Implications for Medicine References PRU-net: An U-net Model with Pyramid Pooling and Residual Block for WMH Segmentation 1 Introduction 2 Related Work 3 Method 3.1 Work Flow 3.2 U-net Based Fully Convolutional Neural Network 3.3 Pyramid Pooling Block 3.4 Residual Connection Block 4 Experiment 4.1 Datasets and Preprocessing 4.2 Experimental Setup 4.3 Evaluation Criteria 4.4 Comparison of Different Models 4.5 Comparison with Existing Approaches 5 Discussion References Two-Way Perceived Color Difference Saliency Algorithm for Image Segmentation of Port Wine Stains 1 Introduction 1.1 A Subsection Sample 2 Principle of Proposed Algorithm 2.1 Pre-processing 2.2 Image Segmentation 2.3 Post-processing 3 Data Sources 4 Results 5 Discussion 6 Conclusion References A New Pathway to Explore Reliable Biomarkers by Detecting Typical Patients with Mental Disorders 1 Introduction 2 Methods 2.1 Data and Neuroimaging Measures 2.2 Overview of Our Method 2.3 Detection of Typical Subjects 2.4 Evaluation 3 Results 3.1 Results of Study 1: Typical Patients Show Significant Group Differences Using Statistical Analyses 3.2 Results of Study 2: Typical Patients Are More Distinguishable Than Whole Subjects Based on Classification Task 3.3 Results of Study 3: Typical Patients Show More Compactness Within Groups and Significant Separation Between Groups Using Clustering and Projection Analyses 4 Conclusion References Activities Prediction of Drug Molecules by Using Automated Model Building with Descriptor Selection 1 Introduction 2 Towards Automated Activities Prediction of Drug Molecules 2.1 Automated Descriptor Selection 2.2 Automated Model Building 3 Experiments 3.1 Dataset 3.2 Experimental Setup 3.3 Experimental Analysis 4 Conclusions References Survival Prediction of Glioma Tumors Using Feature Selection and Linear Regression 1 Introduction 2 Materials and Methods 2.1 Dataset 2.2 Tumor Segmentation and Feature Extraction 2.3 Feature Seletion and Regression 2.4 Implementation Details 3 Results 4 Discussion and Conclusion References AI and Big Data Root Cause Localization from Performance Monitoring Metrics Data with Multidimensional Attributes 1 Introduction 2 Problem Formulation 2.1 Problem Statement 2.2 Challenge 3 Proposed Approach 3.1 Definition of Objective Function 3.2 Explanations of Objective Function 3.3 Heuristic Search Framework 4 Evaluation 4.1 Date Set 4.2 The Effectiveness and Efficiency of Algorithm 5 Conclusion References A Performance Benchmark for Stream Data Storage Systems 1 Introduction 2 Overview of Stream Data Storage Systems 2.1 Typical Application Scenarios 2.2 Requirements 2.3 Critical Technologies 2.4 Typical Systems 3 Design of SSBench 3.1 Architecture and Functions 3.2 Common Read/Write Performance 3.3 Column Read/Write Performance 3.4 Imbalanced Load Performance 3.5 Transactional Load Performance 4 Performance Evaluation of Typical Systems 4.1 Common Read/Write Performance 4.2 Column Stream 4.3 Load Balance 4.4 Distributed Transactions 5 Related Works 5.1 Benchmarks for Storage Systems 5.2 Benchmarks for Data Processing Systems 6 Conclusion References Failure Characterization Based on LSTM Networks for Bluegene/L System Logs 1 Introduction 2 Methodology 2.1 Log Preprocessing 2.2 Vectorization 2.3 Model Training 2.4 Failure Rules Mining 3 Experiments and Evaluations 4 Related Work 5 Conclusion and Future Work References Traffic Crowd Congested Scene Recognition Based on Dilated Convolution Network 1 Introduction 2 Related Work 2.1 Congested Scene Recognition Based on Sensed Data 2.2 Dilated Convolutional Neural Networks 2.3 Limitations of the State-of-the-art Approaches 3 Crowd Congestion Scene Detection Based on Two-Column Very Deep Learning 3.1 Dilated Convolution on Two-Column Network 3.2 Proposed Crowd Congestion Detection Framework 3.3 Constructing Two-Column Dilated Convolutional Network Architecture 3.4 Learning Crowd Congestion Scene Recognition Model Based on Two-Column Dilated Convolution Network 4 Experiment 4.1 Dataset and Metrics 4.2 Training Details 4.3 Comparison Results 5 Conclusion References Failure Prediction for Large-Scale Clusters Logs via Mining Frequent Patterns 1 Introduction 2 Terminology 3 Methodology 3.1 Failure Prediction Framework 3.2 Construct Event Transactions 3.3 Construct Event Sequence Transactions 3.4 Frequent Event Sequences Mining 3.5 Building Failure Rules Library 3.6 Online Failure Prediction 4 Experiments and Evaluations 4.1 Experiment Settings 4.2 Log Characteristics Analysis 4.3 Event Sequence Transactions 4.4 Rules Mining Results 4.5 Evaluation of Failure Predication 5 Related Work 6 Conclusion and Future Works References FLBench: A Benchmark Suite for Federated Learning 1 Introduction 2 Related Work 2.1 Federated Learning 2.2 Benchmarks 3 FLBench Methodology and Design 3.1 FLBench Methodology 3.2 FLBench Design 3.3 FLBench Implementation 4 Conclusion References Fake News Detection Using Knowledge Vector 1 Introduction 2 Related Work 3 Methodology 3.1 Overview 3.2 Extract Event Triple 3.3 Fuse Word2vec and TransE 3.4 Detect Fake News 4 Experiment 4.1 Dateset 4.2 Experimental Setup 5 Conclusion References A Reconfigurable Electrical Circuit Auto-Processing Method for Direct Electromagnetic Inversion 1 Introduction 2 Principle Design 3 Method Demonstration 4 Evaluation and Discussion 4.1 Topological Determination 4.2 Convergence Performance 4.3 Inversion with Admittance 4.4 Complexity Analysis 5 Conclusion References Implementing Natural Language Processes to Natural Language Programming 1 Introduction 2 Brief Natural Language Process 3 Implementing Natural Language Programming 3.1 Sentence Breaker 3.2 Realization of Loop Finder 3.3 Results Display 3.4 Testing 4 Conclusion References AI and Block Chain LSO: A Dynamic and Scalable Blockchain Structuring Framework 1 Introduction 1.1 A Subsection Sample 2 Related Work 2.1 ChainNet 2.2 ABC/TBC Architecture for Scalability and Privacy 2.3 Blockchain Oracles 2.4 Event-Driven Architecture 3 LSO System Structuring Framework 3.1 LSO System Framework 3.2 Collaboration Layer to Support Registration 3.3 Multi-level CL Network 3.4 Dynamic Trust Evaluation 3.5 Event-Driven Architecture (EDA) 3.6 Oracle Machine Operation 4 Applications 4.1 BDL System in LSO 5 Conclusion References CISV: A Cross-Blockchain Information Synchronization and Verification Mode 1 Introduction 2 Related Work 2.1 Blockchain Underlying Storage Mechanism 2.2 Blockchain Interoperability 3 Cross-Blockchain Information Synchronization and Verification 3.1 Definitions 3.2 Cross-Chain Information Synchronization (CIS) 3.3 Cross-Chain Information Verification (CIV) 4 Experiments and Analysis 4.1 On-Chain Data Processing 4.2 Blockchain Storage Performance Test 4.3 Cross-Chain Information Synchronization 4.4 Cross-Chain Information Verification 5 Conclusion References A Formal Process Virtual Machine for EOS-Based Smart Contract Security Verification 1 Introduction 2 Related Work 3 Foundational Concepts 4 Overview of FSPVM-EOS 4.1 Architecture 4.2 Formal Memory Model with Multi-level Table Structure 4.3 Formal Intermediate Specification Language 4.4 Formal Executable Definitional Interpreter for EOS Verification 5 Case Study 6 Conclusion References AVEI: A Scientific Data Sharing Framework Based on Blockchain 1 Introduction 2 Blockchain and Scientific Data Sharing 2.1 Blockchain Technology Overview 2.2 Practical Dilemmas of Scientific Data Sharing 2.3 Coupling Between Blockchain and Scientific Data Sharing 3 Construction of Scientific Data Sharing Framework Based on Blockchain 3.1 Overall Framework Construct 3.2 User Identity Role Authenticate Process 3.3 Data Verify Process 3.4 Data Exchange Process 3.5 Incentive System 4 Performance Analysis of AVEI 4.1 Data Quality Performance Analysis 4.2 Data Security Performance Analysis 4.3 Sharing Effect Performance Analysis 5 Discussion and Conclusion 5.1 Data Quality Performance Analysis References SCT-CC: A Supply Chain Traceability System Based on Cross-chain Technology of Blockchain 1 Introduction 2 Related Work 2.1 Blockchain Technology 2.2 Blockchain in Supply Chain 2.3 Cross-chain Technology 3 Design and Implementation of SCT-CC 3.1 System Architecture 3.2 Smart Contract 3.3 Cross-chain Mechanism 4 Experiment Analysis 4.1 Experimental Environment 4.2 Performance Analysis 5 Conclusion References Game-Theoretic Analysis on CBDC Adoption 1 Introduction 2 Related Work 3 Simple Game-Theoretic Model 4 Detailed Game-Theoretic Model 4.1 Game-Theoretic Settings of the Model 4.2 Construction of Payoff Functions 5 Implementation and Experiments 6 Concluding Remarks and Future Work References Design of Experiment Management System for Stability Control System Based on Blockchain Technology 1 Introduction 1.1 A Subsection Sample 2 Necessity and Feasibility of Blockchain Technology Application on SCS Experiment Management 2.1 Brief Introduction of Blockchain Technology 2.2 Necessity Analysis 2.3 Feasibility Analysis 3 Design of Stability Control Experiment Management System Based on Blockchain 3.1 Definition of Related Concepts 3.2 System Architecture Design 3.3 Analysis of System Operation Process 4 Analysis of System Operation Flow Based on Actual Application 5 Conclusion References A Cross-chain Gateway for Efficient Supply Chain Data Management 1 Introduction 1.1 Background and Significance 1.2 Research Contents 2 Related Work 3 Our Solution 3.1 The Cross-chain Framework of IOTA and Fabric 3.2 The Cross-chain Gateway 3.3 The Cross-chain Data Management Scheme 4 Scheme Validation 4.1 Performance Testing and Analysis 4.2 System Implementation 5 Conclusions References AI and Education Technology Automatic Essay Scoring Model Based on Multi-channel CNN and LSTM 1 Introduction 2 MCNN-LSTM Model 2.1 Embedding-Dense Layer 2.2 MCNN Layer 2.3 LSTM 2.4 Objective and Training 3 Experiment and Analysis 3.1 Dataset 3.2 Evaluation Metrics 3.3 Parameter Settings 3.4 Experimental Results and Discussion 4 Conclusion References Research on Knowledge Graph in Education Field from the Perspective of Knowledge Graph 1 Introduction 2 Data Source and Processing 2.1 Data Source 2.2 Data Processing 2.3 Graph Construction 3 Research Results and Analysis 3.1 Annual Trends of Literature 3.2 Distribution of Journals 3.3 Highly Cited Literature 3.4 Main Research Institutions and Cooperation 3.5 Analysis of Core Authors and Their Cooperation 3.6 Distribution of Data Sources 3.7 Distribution of Data Processing Tools 4 Conclusion and Thinking References Course Evaluation Analysis Based on Data Mining and AHP: A Case Study of Python Courses on MOOC of Chinese Universities 1 Introduction 2 Related Work 3 Evaluation Method and Process 3.1 Data Acquisition 3.2 Course Evaluation 3.3 Comment Analysis and Results 4 Conclusion References Process-Oriented Definition of Evaluation Indicators, Learning Behavior Collection and Analysis: A Case Study 1 Introduction 2 Related Work 3 System of Learning Evaluation Indicators 3.1 Overall Definition 3.2 Detailed Definition 4 Preliminary Knowledge 4.1 KFCoding 4.2 LRS 4.3 Experience API 5 Implementation of Indicators 6 Display and Analysis of Learning Behavior Data 7 Summary and Future Work References The Reform and Construction of Computer Essential Courses in New Liberal Arts Aiming at Improving Data Literacy 1 Introduction 2 Preparation for Teaching Reform 3 Objectives of Teaching Reform 3.1 Improve Data Literacy 3.2 Enhance Social Competitiveness 3.3 Improve Comprehensive Creativity 4 Problems in Teaching Reform 4.1 The Course is Difficult and the Students’ Foundation is Weak 4.2 High Hardware Requirements and Difficult to Achieve the Goal 5 Principles of Teaching Reform 6 Specific Contents of the Reform 6.1 New Liberal Arts Curriculum System 6.2 Content and Form Innovation 7 Achievements of Reform Practice 7.1 Textbook Achievements 7.2 Resource Outcomes 7.3 Student Evaluation 7.4 Students’ Works 8 Conclusion References Entity Coreference Resolution for Syllabus via Graph Neural Network 1 Introduction 2 Methodology 2.1 Input of the Proposed GNN 2.2 Graph Convolutional Neural Network 3 Results and Discussion 3.1 Dataset 3.2 The Setting of the Hyperparamete 3.3 Experiments 3.4 Analysis 4 Conclusion References An Exploration of the Ecosystem of General Education in Programming 1 Introduction 2 Background 2.1 Challenges 2.2 Related Work 3 The Ecosystem of General Education in Programming 3.1 Learning Support Technologies 3.2 Learning Resources 3.3 Instructional Design 3.4 Learning Methods 3.5 Subjects 4 Building the Ecosystem 4.1 Teaching Content 4.2 Question Back Construction 4.3 Deep Teaching 4.4 Evaluation System 5 Experiment 5.1 Classroom Teaching 5.2 Programming Practices 5.3 Teaching Results 5.4 Lessons Learned 6 Conclusions References The New Theory of Learning in the Era of Educational Information 2.0—Connected Constructivism 1 Introduction 2 Constructivism 2.1 The Background of Constructivism 2.2 The Main Points of Constructivism 2.3 Evaluation 3 Connectivism 3.1 The Background of Connectivism Theory 3.2 The Main Points of Connectivism 3.3 Evaluation 4 New View-Connected Constructivism 4.1 Connected Constructivism View of Knowledge 4.2 Connected Constructivism View of Learning 4.3 Connected Constructivism View of Teaching 4.4 Connected Constructivism View of Students 4.5 Connected Constructivism View of Teachers 5 Conclusion References AI and Financial Technology A Stock Index Prediction Method and Trading Strategy Based on the Combination of Lasso-Grid Search-Random Forest 1 Introduction 2 Construction of L-GSRF Model 2.1 Theoretical Model of L-GSRF 2.2 Implementation of L-GSRF Model 3 Data Collection and Processing 3.1 Data Collection 3.2 Data Processing 4 L-GSRF Model Experimental Results and Analysis 4.1 Lasso Regression Result 4.2 L-GSRF Model Prediction Results 5 Trading Strategy Based on L-GSRF Model 6 Research Conclusions and Reflections References Dynamic Copula Analysis of the Effect of COVID-19 Pandemic on Global Banking Systemic Risk 1 Introduction 2 Methodology 2.1 Truncated D-vine Copula 2.2 Dynamic Mixture of Time-Varying Copulas 3 Data Description and Marginal Distribution 4 Empirical Analysis 4.1 Systemic Risk Measures 4.2 Systemic Risk Level Analysis 4.3 Systemic Risk Contribution Analysis 5 Conclusion References Real-Time Order Scheduling in Credit Factories: A Multi-agent Reinforcement Learning Approach 1 Introduction 2 Related Work 2.1 Single-Agent Reinforcement Learning 2.2 Multi-agent Reinforcement Learning 3 Problem Formulation 4 Methodology 4.1 Framework of MARL Based Order Scheduling 4.2 Reward Calculation 4.3 State Generation 5 Numerical Experiments 5.1 Virtual Credit Factory 5.2 Performance Measures and Baseline Algorithms 5.3 Experimental Settings 5.4 Performance Measures and Baseline Algorithms 5.5 Robustness Check 5.6 The Results of Online A/B Tests 6 Conclusion References Predicting Digital Currency Price Using Broad Learning System and Genetic Algorithm 1 Introduction 2 Literature Review 2.1 Digital Currency Price Prediction 2.2 Broad Learning System 2.3 Summary and Importance of the Proposed Work 3 A Prediction Model for Digital Currency Price 3.1 Price Prediction Based on the Broad Learning System 3.2 Model Optimization Based on the Genetic Algorithm 3.3 Model Training and Evaluation 4 Experiment Result and Analysis 4.1 Experimental Environment 4.2 Data Description and Analysis 4.3 Digital Currency Prediction Results and Analysis 5 Summary and Future Work References Selective Multi-source Transfer Learning with Wasserstein Domain Distance for Financial Fraud Detection 1 Introduction 2 Related Work 2.1 Financial Fraud Detection 2.2 Transfer Learning 3 Methodology 3.1 Problem Formulation 3.2 Self-supervised Domain Distance Learning Module 3.3 Single-Source-Single-Target Transfer Module 3.4 Aggregation Module 4 Experiment 4.1 Experiment Settings 4.2 Results on Domain Relationships 4.3 Comparison Results 4.4 W-Distance Vs. SSST Transfer 5 Conclusion References Author Index