Intelligent Computers, Algorithms, and Applications: Third BenchCouncil International Symposium, IC 2023, Sanya, China, December 36, 2023, Revised ... in Computer and Information Science, 2036)

This book constitutes the proceedings of the Third BenchCouncil International Symposium on Intelligent Computers, Algorithms, and Applications, IC 2023, which took place in Sanya, China, in December 2023. The 18 full papers and 8 short papers included in this book were carefully reviewed and selected from 50 submissions. They were organized in topical sections as follows: AI Algorithms and Systems; AI for Ocean science and engineering; AI in finance; AI for education; AI for materials science and engineering; AI for medicine; AI for civil aviation; AI for high energy physics; AI for law. IC 2023: Provide a Pioneering AI Technology Roadmap Organization Contents AI Algorithms and Systems Efficient and Scalable Kernel Matrix Approximations Using Hierarchical Decomposition 1 Introduction 1.1 Motivation 1.2 Proposed Approach 1.3 Related Work and Contributions 2 Methods 2.1 Diffusion Maps 2.2 Implicitly Restarted Arnoldi Method 3 Implementation 3.1 Integration of datafold and GOFMM 3.2 LinearOperator 4 Results 4.1 Eigenvalue and Eigenvector Computations 4.2 Scaling 5 Conclusion References Second-Order Gradient Loss Guided Single-Image Super-Resolution 1 Introduction 2 Related Works 2.1 Single Image Super-Resolution Methods 2.2 Gradient Guided Super-Resolution Methods 3 Our Method 3.1 Problem Definition 3.2 Second-Order Gradient Loss 4 Experiments 4.1 DataSets and Metrics 4.2 Implementation Details 4.3 Quantitative and Qualitative Comparisons 5 Conclusion References The Implementation and Optimization of FFT Calculation Based on the MT-3000 Chip 1 Introduction 2 Chip Overall Structure 3 FFT Algorithm Implementation 3.1 Base 2FFT 3.2 Matrix Line and Column FFT 4 Optimization of FFT Algorithm 4.1 Data Processing 4.2 DMA Dual Channel Transmission 4.3 Vectorization 4.4 Double Buffer Transmission 4.5 MPI 4.6 Linear Assembly 5 Experiment and Analysis 5.1 Lab Environment 5.2 Correctness Analysis 5.3 Performance Analysis 6 Conclusion References EDFI: Endogenous Database Fault Injection with a Fine-Grained and Controllable Method 1 Introduction 2 Background and Related Work 3 Overview of EDFI 4 EDFI's Design and Implementation 4.1 System Initialization 4.2 Fault Injection Point Selection 4.3 Fault Injection Precondition Generation 4.4 Fault Injection Policy Generation 4.5 Fault Recovery 5 Evaluation 5.1 Experimental Setup 5.2 Fault Injection Effectiveness of EDFI 5.3 Further Validate the Effectiveness of EDFI Through Anomaly Detection Algorithms 5.4 Sensitivity Analysis of Anomaly Detection Algorithms 6 Limitations 7 Conclusion References AI for Ocean Science and Engineering Diffusion Probabilistic Models for Underwater Image Super-Resolution 1 Introduction 2 Related Work 2.1 SISR for Underwater Imagery 2.2 Diffusion Probabilistic Models 3 Method 3.1 Diffusion Model 3.2 DiffUSR 4 Experiments 4.1 Implementation Details 4.2 Experimental Analysis 4.3 Ablation Study 5 Conclusion References Classification Method for Ship-Radiated Noise Based on Joint Feature Extraction 1 Introduction 2 Theoretical Work 2.1 Wavelet Transform 2.2 Mel Spectrum Features 2.3 Convolutional Neural Networks 3 Experiments 3.1 Dataset Introduction 3.2 Joint Feature Extraction 3.3 Convolutional Neural Network Structure 4 Results 5 Conclusion References AI in Finance Forecasting the Price of Bitcoin Using an Explainable CNN-LSTM Model 1 Introduction 2 Methodology 2.1 Convolution Neural Network (CNN) 2.2 Long Short-Term Memory (LSTM) 2.3 Shapley Additive Explanations (SHAP) 3 Experiment 4 Conclusion References Augmenting Bankruptcy Prediction Using Reported Behavior of Corporate Restructuring 1 Introduction 2 Related Works 3 Methodology 3.1 Conceptual Framework 3.2 Variables and Data 3.3 Experimental Setup 4 Results and Discussion 4.1 Features Evaluation 4.2 Ablation Experimental Results 4.3 Performance About Covid Period 5 Conclusion A Appendix A References AI for Education A New Dataset and Method for Creativity Assessment Using the Alternate Uses Task 1 Introduction 2 The Cambridge AUT Dataset 2.1 Data Collection 2.2 Annotation 3 Semantic Models 3.1 Semantic Distance 3.2 Confirmatory Factor Analysis 4 Binary Classification Models 4.1 Fine-Tuned Models 4.2 Results 4.3 A Case Study with New AUT Prompts 4.4 Comparison with ChatGPT Predictions 5 Conclusions 6 Limitations A The Instructions Used for the AUT B Detailed CFA Results C Cross Validation Results D Model Performance on the Dataset from ch9beatyspsetalsps2018 E ChatGPT Prompt F ChatGPT Classification Results References AI for Materials Science and Engineering Convolutional Graph Neural Networks for Predicting Enthalpy of Formation in Intermetallic Compounds Using Continuous Filter Convolutional Layers 1 Introduction 2 Methodologies 2.1 Graph Construction 2.2 Continuous Filter Convolutional Layer in Graph Convolutional Network 2.3 Graph Neural Network Architecture 3 Experiment 3.1 Dataset 3.2 Implementation Details 4 Results and Discussion 5 Conclusion References Predicting Li Transport Activation Energy with Graph Convolutional Neural Network 1 Introduction 2 Method 2.1 IN Calculation Method 2.2 IN Representation Method 2.3 Construction of GCN-Based Activation Energy Prediction Model 3 Experiments 3.1 Dataset 3.2 Experimental Setup 3.3 Experimental Results 3.4 Parameter Sensitivity Analysis 4 Conclusion References AI for Medicine KGCN-DDA: A Knowledge Graph Based GCN Method for Drug-Disease Association Prediction 1 Introduction 2 Methods and Materials 2.1 Dataset 2.2 Drug–Disease Association Prediction Based on Knowledge Graph and GCN 3 Results and Discussion 3.1 Performances and Comparison with State-of-the-Art Methods 3.2 Case Study 4 Conclusions References Machine Learning for Time-to-Event Prediction and Survival Clustering: A Review from Statistics to Deep Neural Networks 1 Introduction 2 Preliminaries 3 Time to Even Prediction 3.1 Statistics Based Methods 3.2 Traditional Machine Learning Based Methods 3.3 Neural Network Based Methods 4 Cluster Based Risk Profile 4.1 Statistics Based Methods 4.2 Traditional Machine Learning Based Methods 4.3 Neural Network Based Methods 5 Conclusions and Future Directions References Label-Independent Information Compression for Skin Diseases Recognition 1 Introduction 2 Method 2.1 Information Bottleneck 2.2 Hilbert-Schmidt Independence Criterion 2.3 The Proposed CIH Algorithm 3 Experiments 3.1 Datasets 3.2 Implementation Details 3.3 Results 4 Conclution References AI for Civil Aviation 3D Approach Trajectory Optimization Based on Combined Intelligence Algorithms 1 Introduction 2 Thunderstorm Movement Prediction 3 Trajectory Prediction Algorithm of Approaching Aircraft 3.1 Artificial Potential Field Algorithm 3.2 Rapidly-Exploring Random Tree Algorithm 3.3 The Combined Algorithm 4 Case Study References A-SMGCS: Innovation, Applications, and Future Prospects of Modern Aviation Ground Movement Management System 1 Introduction 2 A-SMGCS Overview 2.1 Concept of A-SMGCS 2.2 Technological Evolution of A-SMGCS 3 Components of A-SMGCS 4 Key Technologies of A-SMGCS 4.1 Data Fusion and Processing 4.2 Surveillance Data Analysis and Positioning 4.3 Flight Data Processing and Matching 4.4 Path Planning 4.5 Conflict Detection and Alerting 4.6 Human-Machine Interface and Operational Support 4.7 Digital Pre-departure Clearance System (DCL) 5 Application and Benefits of A-SMGCS 6 Challenges and Future Development 7 Conclusion References AI for High Energy Physics An Intelligent Image Segmentation Annotation Method Based on Segment Anything Model 1 Introduction 2 Method 2.1 Segment Anything Model 2.2 SAM Worker 2.3 GUI 3 Usage and Application Cases 3.1 API Usage 3.2 GUI Usage 3.3 Use Cases 3.4 Comparison 4 Conclusion References ParticleNet for Jet Tagging in Particle Physics on FPGA 1 Introduction 2 Methodology 2.1 Operator Fusion 2.2 Model Quantization 3 HardWare Design 3.1 Overall System Architecture 3.2 Implementation Design 4 Results and Discussion 4.1 Experiments Setup 4.2 Results 4.3 Discussion References Application of Graph Neural Networks in Dark Photon Search with Visible Decays at Future Beam Dump Experiment 1 Introduction 2 Experimental Setup and Simulation Framework 2.1 Experimental Setup 2.2 Event Simulation 3 Analysis Strategy and Tracking Reconstruction 3.1 GNN-Based Tracking Reconstruction 3.2 Dark Photon Search in Visible Decay Mode 4 Conclusion and Outlook References Neutrino Reconstruction in TRIDENT Based on Graph Neural Network 1 Introduction 2 Event Simulation 3 Network Architecture 4 Results 4.1 Shower-Like Event Reconstruction 4.2 Track-Like Event Reconstruction 5 Summary References Charged Particle Reconstruction for Future High Energy Colliders with Quantum Approximate Optimization Algorithm 1 Introduction 2 Track Reconstruction 2.1 Current Classical Benchmarks 2.2 Quantum Approach 3 Datasets 4 Methodology 4.1 Optimization of QAOA Implementation 4.2 Sub-QUBO Method 5 Results and Discussions References AI for Law A Levy Scheme for User-Generated-Content Platforms and Its Implication for Generative AI Providers 1 Introduction 2 The UGC Dilemma Due to the Lack of Regulation on Intermediaries 3 The Intermediary-Oriented Approach Underlying Copyright Law History 3.1 Producers and Distributors in Copyright Law 3.2 The Safe Harbor Doctrine Exempting the Liability of Intermediaries 4 Imposing Levies on UGC Platforms 4.1 Levy Rather Than Compulsory License 4.2 Levying UGC Platforms with Non-commercial UGC Creation Exempted 5 The Justification of the Non-commercial UGC Creation Levy 5.1 Tailored to the Democratization of Re-Creating Content and Making Content Available to the Public 5.2 Fair Remuneration to Copyright Owners 5.3 Privacy Concern 5.4 Cultural Promotion 6 Concluding Remarks References Moving Beyond Text: Multi-modal Expansion of the Toulmin Model for Enhanced AI Legal Reasoning 1 Introduction 2 The Toulmin Model and Legal Reasoning 2.1 The Introduction of the Toulmin Model 2.2 Challenges Faced in the Context of Multi-modal Digital Data 3 Expansion of the Toulmin Model 3.1 The Theoretical Basis for Multi-modal Integration 3.2 The Multi-modal Expansion of the Toulmin Model 4 Conclusion: The Future of Legal Argumentation in the Age of AI References The Worldwide Contradiction of the GAI Regulatory Theory Paradigm and China's Response: Focus on the Theories of Normative Models and Regulatory Systems 1 Introduction 2 Sporadic Governance: Attribution of Worldwide Problems and Dilemmas 2.1 The Limitations of the Single Planning Governance Paradigm 2.2 The Dilemma of Theoretical Implementation in the Attempt to Update the Regulatory System 3 Dynamic Integration: China's Response and Development Direction 3.1 Mesh Specification Under System Optimization 3.2 The Balance of Governance-Type Supervision Under Antinomy 4 Postscript: GAI, Is It a New Era or a Phantom? References Intelligent Forecasting of Trademark Registration Appeal with TF-IDF and XGBoost 1 Introduction 2 Related Work 3 Predictive Modeling 3.1 Data Soures and Model Construction Ideas 3.2 TF-IDF Algorithm 3.3 XGBoost Model 4 Experiments and Analysis of Results 4.1 Experimental Procedure 4.2 Parameter Tuning 4.3 Feature Analysis 4.4 Experimental Results 5 Limitations and Future Work References Review of Big Data Evidence in Criminal Proceedings: Basis of Academic Theory, Practical Pattern and Mode Selection 1 Introduction 2 Practice Pattern of Big Data Evidence Review 3 Multidimensional Flaws of Big Data Evidence Review 3.1 Defects in the Technical Dimension of Big Data Evidence Review 3.2 Defects in the Legal Dimension of Big Data Evidence Review 4 Mode Selection for Big Data Evidence Review 4.1 Regulation of Litigation Procedure: The Path Correction of Big Data Evidence Transmission 4.2 Optimization of Proof Mode: Probabilistic Analysis of Evidence From Big Data 4.3 Digital Rights Protection: The Construction of Illegal Exclusion Rules for Big Data Evidence 5 Conclusion References Abstracts Application of Machine Learning-Based Neural Networks in Positron Annihilation Spectroscopy Data Analysis References Machine Learning Techniques for Automatic Detection of ULF Waves 1 Main Text References Reconstruction of Unstable Heavy Particles Using Deep Symmetry-Preserving Attention Networks References Interpretable Prediction of Commercial Flight Delay Duration Based on Machine Learning Methods Semantic Retrieval of Mars Data Using Contrastive Learning and Convolutional Neural Network 1 Introduction References Author Index