Advanced Data Mining and Applications

Name: Advanced Data Mining and Applications | 17th International Conference, ADMA 2021, Sydney, NSW, Australia, February 2-4, 2022, Proceedings, Part I
Brand: Springer
Price: 85.59 EUR
Availability: OnlineOnly

17th International Conference, ADMA 2021, Sydney, NSW, Australia, February 2-4, 2022, Proceedings, Part I

Bohan Li Lin Yue Jing Jiang Weitong Chen Xue Li Guodong Long Fei Fang Han Yu(Herausgeber*in)

Springer (Verlag)

Erschienen am 31. Januar 2022

XVII, 439 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-030-95405-5 (ISBN)

85,59 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Intro
Preface
Organization
Contents - Part I
Contents - Part II
Healthcare
Deep Learning Based Cardiac Phase Detection Using Echocardiography Imaging
1 Introduction
2 Related Work
3 Methodology
3.1 Phase I: Model Training
3.2 Phase II: Model Testing
4 Experiments
4.1 Datasets
4.2 Competing Approaches
4.3 Parameter Configuration
4.4 Hardware and Software Configuration
5 Evaluation
5.1 Evaluation Metrics
5.2 Comparison Among Competing Approaches
5.3 Effect of Image Preprocessing
5.4 Evaluation of Parameter Sensitivity
5.5 Performance Analysis of the Custom Loss Function
5.6 Evidence of Generalization of DeepPhase
6 Conclusion
References
An Empirical Study on Human Flying Imagery Using EEG
1 Introduction
2 Method
2.1 Experiment and EEG Recording
2.2 Classification and Feature Analysis
3 Results
3.1 Overall Classification Results
3.2 Frequency Band Specific Classification Results
3.3 Time Window Specific Classification Results
3.4 Time-frequency Specific Classification Results
3.5 EEG Activity Patterns in Most Significant Time-frequency Bin
4 Discussion
5 Conclusion and Future Work
References
Network Graph Analysis of Hospital and Health Services Functional Structures
1 Introduction
2 Subjects and Methods
3 Network Graphs
4 Results
5 Discussion
6 Conclusion and Future Work
References
Feature Selection in Gene Expression Profile Employing Relevancy and Redundancy Measures and Binary Whale Optimization Algorithm (BWOA)
1 Introduction
1.1 Objective and Contributions
2 Related Work
3 Proposed Methodology
3.1 Feature Scaling
3.2 Phase 1 Feature Selection: Relevance Analysis
3.3 Phase 2 Feature Selection: Redundancy Analysis
3.4 Phase 3 Feature Selection: Meta-heuristic Optimization
3.5 Binary Whale Optimization Algorithm (BWOA)
3.6 BWOA for Gene Selection
3.7 Classification
4 Datasets and Baselines
4.1 Gene Expression Datasets
4.2 Performance Metrics
4.3 Baseline Methods
5 Results and Discussion
6 Summary and Conclusions
References
Hand Bone Age Estimation Using Deep Convolutional Neural Networks
1 Introduction
2 Background and Related Works
3 Materials and Methods
3.1 Normalization
3.2 Hand Detection
3.3 Vision Pipeline
3.4 Proposed Bone Age Prediction Model
3.5 Dataset
4 Experimental Results and Discussion
5 Conclusion
References
An Interpretable Machine Learning Approach for Predicting Hospital Length of Stay and Readmission
1 Introduction
2 Methods
3 Results and Discussion
4 Conclusion
References
STCT: Spatial-Temporal Conv-Transformer Network for Cardiac Arrhythmias Recognition
1 Introduction
2 Related Works
2.1 Diagnosis of Cardiac Arrhythmias
2.2 Deep Learning-Based Cardiac Arrhythmias Diagnose
3 Methodology
3.1 Data Segmentation
3.2 the Proposed Model
4 Experiment
4.1 Datasets and Model Implementation
4.2 Comparison Model
4.3 Experimental Results
5 Conclusion
References
Education
Augmenting Personalized Question Recommendation with Hierarchical Information for Online Test Platform
1 Introduction
2 Preliminary
2.1 Definitions and Problem Statement
2.2 Framework Overview
3 Method
3.1 Incorporating the Student and Question Hierarchical Information
3.2 Student Performance Predicting
3.3 The Framework APQR
4 Experiments
4.1 Online Test Dataset
4.2 Evaluation Metric
4.3 Baseline Algorithms
4.4 Overall Performance
4.5 Parameter Analysis
5 Related Work
5.1 Recommender System
5.2 Student Performance Modeling
6 Conclusion
References
Smart Online Exam Proctoring Assist for Cheating Detection
1 Intruduction
2 Related Work
3 Proposed Technique Highlight
3.1 Problem Statement
3.2 High Level Architecture
4 Proposed Work Details
4.1 Exam Recording
4.2 Video Characteristic Analysis
4.3 Videos Transformed to Feature Vector
4.4 Data Uniforming by Video Length Equalizing
4.5 Training
5 Experiments
5.1 Dataset
5.2 Competing Approaches
5.3 Parameters, Hardware and Software
5.4 Evaluation
6 Conclusion
References
Design and Development of Real-Time Barrage System for College Class
1 Introduction
2 System Design
2.1 System Framework and Function Design
2.2 System Flow Design
3 Analysis of Sensitive Word Filtering Algorithm
4 System Realization
4.1 PC Function Realization
4.2 Mobile Function Realization
4.3 Server-Side Function Realization
5 Conclusion
References
Recommendation for Higher Education Candidates: A Case Study on Engineering Programs
1 Introduction
2 Related Work
3 ESTHER
3.1 Students Profiler
3.2 Programs Recommender
3.3 System Dependencies and Limitations
4 Case Study
4.1 Students Profiler
4.2 Programs Recommender
4.3 ESTHER Overview
5 Conclusions
References
Web Application
UQ-AAS21: A Comprehensive Dataset of Amazon Alexa Skills
1 Introduction
2 Background and Related Work
2.1 Background
2.2 Related Work
3 The UQ-AAS21 Dataset
3.1 Data Scraping
3.2 Data Processing
3.3 Dataset Features
4 Preliminary Studies Based on UQ-AAS21 Datasets
4.1 Demographic Study
4.2 Analysis of Privacy Policy and Term of Use Document
5 Potential Usage of UQ-AAS21
6 Conclusion
References
Are Rumors Always False?: Understanding Rumors Across Domains, Queries, and Ratings
1 Introduction
2 Background
2.1 Detecting Rumors on the Web
2.2 Actions Against Detected Rumors
2.3 Research Gap
3 Research Questions
4 Methodology
4.1 Data Collection
5 Empirical Analyses and Findings
5.1 What Are the Rumors About?
5.2 Where Do the Rumors Come From?
5.3 Who Contribute to Rumors?
5.4 When Are the Rumors Reported?
5.5 How Do Rumors Propagate?
6 Discussion
6.1 Key Findings
6.2 Implications for Public Trust and Explainable Rumor Detection
6.3 Limitations and Future Work
7 Conclusion
References
A Green Pipeline for Out-of-Domain Public Sentiment Analysis
1 Introduction
2 Related Work
3 Methodology
3.1 Problem Definition
3.2 Pre-trained Transformer Encoder
3.3 Pipeline Sentiment Analysis Model
4 Experiments
4.1 Experimental Setups
4.2 Sentiment Analysis Evaluation
4.3 Performance of Sub-models
4.4 Analysis and Case Study
5 Conclusion
References
Profiling Fake News: Learning the Semantics and Characterisation of Misinformation
1 Introduction
2 Experimental Dataset
2.1 Data Pre-processing
3 Proposed Solution Approach
3.1 Features Extraction and Selection
3.2 Classification Models
4 Experimental Results
5 Conclusion and Future Work
References
Mining Social Networks for Dissemination of Fake News Using Continuous Opinion-Based Hybrid Model
1 Introduction
2 Proposed Model
2.1 Initialization
2.2 Propagation
3 Results and Discussions
4 Conclusion
References
Predicting Network Threat Events Using HMM Ensembles
1 Introduction
2 Background and Related Work
3 Ensemble of Hidden Markov Models
3.1 Hidden Markov Model Structure
3.2 Event Sequence Clustering
3.3 Ensemble Creation and Prediction Methods
4 Data Set
5 Evaluation
6 Conclusion
References
On-device Application
Group Trip Planning Queries on Road Networks Using Geo-Tagged Textual Information
1 Introduction
2 Background and Problem Definition
3 Proposed Solution Methodologies
3.1 Brute Force Approach with Precomputed Distance
3.2 Group Nearest Neighbor (GNN) to Compute GTP Queries
3.3 Using R-trees to Compute GTP Queries
4 Experimental Evaluation
5 Conclusion and Future Direction
References
Deep Reinforcement Learning Based Iterative Participant Selection Method for Industrial IoT Big Data Mobile Crowdsourcing
1 Introduction
2 System Model and Deep Neural Network
2.1 System Model
2.2 Participant Selection Problem
3 System Framework and Deep Q-Network
3.1 System Framework
3.2 Deep Q-Network
4 Evaluation
4.1 Dataset and Experiment Setups
4.2 BaseLine Method
4.3 The Performance Evaluation and Comparison
5 Related Work
6 Conclusion
References
Know Your Limits: Machine Learning with Rejection for Vehicle Engineering
1 Introduction
2 Background
2.1 Vehicle Engineering: The Need for Usage Profiling
2.2 Related Work on Machine Learning with a Reject Option
3 Usage Profiling: Data Science Challenge
4 Our Approach for Vehicle Usage Profiling
4.1 Predictor h
4.2 Rejector r
4.3 Combined Model h'
5 Use-Case: Road-Roughness Analysis
5.1 Data Collection and Preprocessing
5.2 Experimental Methodology
5.3 Results
5.4 Discussion and Lessons Learned
6 Conclusion
References
Towards Generalizable Machinery Prognostics
1 Introduction
2 Background and Related Work
3 A Generic Approach to Incipient Failure Prediction
4 Learning the Prediction Horizon
5 Ablation Study
6 Results and Discussion
7 Conclusion
8 Future Work
References
A Trust Management-Based Route Planning Scheme in LBS Network
1 Introduction
2 Related Work
3 Framework of System
3.1 Definition of Entities
3.2 Blockchain for System
4 Trust Management
4.1 The Management of Certificate
4.2 Malicious Node Identification
5 Route Planning Scheme
5.1 Framework of DTP
5.2 Graph Partitioning
5.3 Distributed Deployment of DTP
5.4 Design Mentality of TORP
6 Experiment
6.1 Experiments Settings and Datasets
6.2 Trust Management Evaluation
6.3 TORP Evaluation
7 Conclusion
References
FreeSee: A Parameter-Independent Pattern-Based Device-Free Human Behaviour Sensing System with Wireless Signals of IoT Devices
1 Introduction
2 Related Works
2.1 Pattern-Based Approaches
2.2 Model-Based Approaches
3 Motivation
3.1 Correlation Between Sensing Accuracy and the Grains of the Wireless Signatures
3.2 Correlation Between Sensing Accuracy and Feature Extraction Parameters
4 Design of FreeSee
4.1 Correlation Between Sensing Accuracy and Classification Parameters
5 Implementation and Experimental Results
5.1 Performance Metrics
5.2 Performance of the Optimized Results
5.3 Comparison with Other Algorithms
5.4 Comparison with the Algorithm Without GA
5.5 Comparison with the Algorithm Without GA
6 Conclusion
References
Others
PS-QMix: A Parallel Learning Framework for Q-Mix Using Parameter Server
1 Introduction
2 Related Works
2.1 Multi-agent Reinforcement Learning
2.2 Distributed Reinforcement Learning
3 The Parameter Server Q-Mix Algorithm
3.1 The Worker
3.2 The Parameter Server
3.3 Main Program and Its Ray Implementation
4 Experiments
5 Conclusion
References
A Comprehensive Feature Importance Evaluation for DDoS Attacks Detection
1 Introduction
2 Related Work
3 Feature Importance Evaluation
3.1 AUC
3.2 DT and RF
3.3 Mutual Information
4 Experiment Results
4.1 Datasets
4.2 Evaluation
4.3 Experiment Results
5 Conclusion
References
Adaptive Fault Resolution for Database Replication Systems
1 Introduction
2 Current Fault Resolution Approaches
3 Adaptive Fault Resolution (FR) Module Design
3.1 Diagnostic Reinforcement Learning (DRL) for FR Module
3.2 System Correction (SC) Module
3.3 Representation and Correlation of Diagnosed Faults to Corrective Actions
3.4 Prioritization of the Software Groups' Action
3.5 Cost Function and Q-Values for FR Module
4 Empirical Analysis
4.1 Test Results
4.2 FR Module - SC's Results
4.3 FR's Efficacy Test Results
5 Conclusion
References
An Adjustable Diversity Metric for Multimodal Multi-objective Evolutionary Algorithms
1 Introduction
2 Related Works
3 Proposed Metric of PSCR
4 Experiment
4.1 Experimental Setting
4.2 Analysis of the Performance Metrics
4.3 Results on Comparison by Different Metrics
4.4 Discuss
5 Conclusion
References
Cybersecurity Analysis via Process Mining: A Systematic Literature Review
1 Introduction
2 Background
3 Related Work
4 Methodology
5 Process Mining Used for Cybersecurity
5.1 Security of Industrial Control Systems
5.2 Security of Smartphones
5.3 Web-Application Security
5.4 Network Traffic Security
5.5 Attack Inspection
5.6 Outlier User Behavior Detection
5.7 Fraud Detection
6 Results
7 Threats to Validity
8 Conclusion
References
Identification of Stock Market Manipulation with Deep Learning
1 Introduction
2 Related Work
3 Stock Market Manipulation Identification
3.1 New Stock Market Data for Anomaly Detection
3.2 Source
3.3 Anomalies
3.4 Approaches Applied
3.5 Environment
4 Results and Discussion
5 Conclusion and Future Work
References
Fuzzy Kolmogorov Complexity Based on Fuzzy Decompression Algorithms and Its Application to Fuzzy Data Mining
1 Background and Main Contribution
1.1 Fuzzy Data and Fuzzy Turing Machines
1.2 Kolmogorov Complexity and Applications to Data Mining
1.3 Our Distinctive Approach to Generic Fuzzy Kolmogorov Complexity and an Application to Fuzzy Data Mining
2 Preparation: Basic Notions and Notation
2.1 Numbers, Strings, and Fuzzy Data
2.2 Generic Fuzzy Strings and Fuzzy Mean Values
2.3 Generic Fuzzy Turing Machine
2.4 Fuzzy Interference of Crisp Computation Paths
3 Generic Fuzzy Kolmogorov Complexity
3.1 Foundations of Fuzzy Kolmogorov Complexity
3.2 An Encoding Scheme of Fuzzy Strings
3.3 Fundamental Properties of FKC
4 Applications of Fuzzy Kolmogorov Complexity to Fuzzy Data Mining
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken