Database Systems for Advanced Applications. DASFAA 2022 International Workshops: BDMS, BDQM, GDMA, IWBT, MAQTDS, and PMBD, Virtual Event, April 11–14, 2022, Proceedings

✍ Scribed by Uday Kiran Rage (editor), Vikram Goyal (editor), P. Krishna Reddy (editor)

• Publisher: Springer
• Year: 2022
• Tongue: English
• Leaves: 446
• Series: Lecture Notes in Computer Science; 13248
• Category: Library

✦ Synopsis

This volume constitutes the papers of several workshops which were held in conjunction with the 27th International Conference on Database Systems for Advanced Applications, DASFAA 2022, held as virtual event in April 2022.

The 30 revised full papers presented in this book were carefully reviewed and selected from 65 submissions.

DASFAA 2022 presents the following five workshops:

· First workshop on Pattern mining and Machine learning in Big complex Databases (PMBD 2021)

· 6th International Workshop on Graph Data Management and Analysis (GDMA 2022)

· First International Workshop on Blockchain Technologies (IWBT2022)

· 8th International Workshop on Big Data Management and Service (BDMS 2022)

· First workshop on Managing Air Quality Through Data Science

· 7th International Workshop on Big Data Quality Management (BDQM 2022).

✦ Table of Contents

Preface
Organization
Contents
PMDB
An Algorithm for Mining Fixed-Length High Utility Itemsets
1 Introduction
2 Background
2.1 Problem Description and Definitions
2.2 Related Work
3 Algorithm HUIKM
3.1 Create a Tree and a Header Table
3.2 Ming HUIK from a Tree
4 Experimental Results
5 Conclusion
References
A Novel Method to Create Synthetic Samples with Autoencoder Multi-layer Extreme Learning Machine
1 Introduction
2 SMOTE Method
3 Proposed AE-MLELM-SynMin Method
3.1 Training AE-MLELM
3.2 Conducting Crossover and Mutation Operations
3.3 Creating Synthetic Samples
4 Experiments
4.1 Experiment Setting
4.2 Information Amount Analysis of SMOTE and AE-MLELM-SynMin
4.3 Comparison Among AE-MLELM-SynMin, SMOTE, Borderline-SMOTE, Random-SMOTE, and SMOTE-IPF
5 Conclusion
References
Pattern Mining: Current Challenges and Opportunities
1 Introduction
2 C1: Mining Patterns in Complex Graph Data
3 C2: Targeted Pattern Mining
4 C3: Repetitive Sequential Pattern Mining
5 C4: Incremental, Stream and Interactive Pattern Mining
6 C5: Heuristic Pattern Mining
7 C6: Mining Interesting Patterns
8 Conclusion
References
Why Not to Trust Big Data: Discussing Statistical Paradoxes
1 Introduction
2 Why Not to Trust on Data Science, AI, ML and Big Data
3 Statistical Paradoxes
3.1 Berkson Paradox
3.2 Yule-Simpson's Paradox
4 Existence of Simpson's Paradox in Big Data
4.1 Datasets
5 Analysis Simpson's Paradox in Real Life: A Case Study
5.1 The Dataset
5.2 Data Analysis
6 Discussion
7 Conclusion
References
Localized Metric Learning for Large Multi-class Extremely Imbalanced Face Database
1 Introduction
2 Metric Learning as an Antidote for the Class Imbalance Problem
3 Localized Metric Learning – The Proposed Approach
3.1 Division of Dataset Into Subsets
3.2 Localized Metric Learning
3.3 Algorithm
4 Results and Discussions
5 Conclusions
References
Top-k Dominating Queries on Incremental Datasets
1 Introduction
2 Literature Review
2.1 Top-k Dominance Query
2.2 Dynamic Update of Data Mining
3 Query Base Preparation
4 Algorithm Description
5 Experiment and Analysis
6 Conclusion
References
IWBT
Collaborative Blockchain Based Distributed Denial of Service Attack Mitigation Approach with IP Reputation System
1 Introduction
2 Theoretical Background
2.1 Blockchain Technology – Brief
2.2 Overview of DDoS Mitigation Approaches and Techniques
2.3 IP Reputation Scheme
3 Problem Description
4 Related Work
5 Research Questions
6 Approach and Next Step
6.1 Proposed Architecture
6.2 Integration of IP Reputation System with Proposed Architecture
6.3 Implementation Strategy
6.4 Advantages of Proposed System Over Existing Implementation
7 Conclusion
References
Model-Driven Development of Distributed Ledger Applications
1 Introduction
1.1 Running Example
2 Background
2.1 Distributed Ledger Technology and Hybrid DLT/DB Applications
2.2 MDD with the Interaction Flow Modeling Language
3 Development of Hybrid DLT/DB Applications
3.1 Requirement Specification
3.2 Data Design
3.3 Interface Design
3.4 Operation Design
3.5 Architecture Design
4 Implementation
5 Related Work
6 Conclusion
References
Towards a Blockchain Solution for  Customs Duty-Related Fraud
1 Introduction
2 Related Literature
3 Types of Blockchains for Customs Enforcement
4 Hyperledger Fabric
4.1 Transaction Overview
4.2 Ensuring Transaction Order
5 Detecting and Mitigating Customs Duty Fraud Using Fabric: Three Scenarios
5.1 Underreporting the Cargo Weight or Quantity
5.2 Relying on a Bill of Lading for Customs Audits
5.3 Misrepresenting the Country of Origin: Shipping Goods to an Intermediate Country to Avoid Import Tariffs
6 Benchmarking Scalability
6.1 Benchmarking Setup
6.2 Performance Benchmarking Results and Analysis
7 Other Implementation Considerations
8 Conclusion
References
Securing Cookies/Sessions Through Non-fungible Tokens
1 Introduction
2 Background
3 Literature Review
3.1 Blockchain
3.2 Cookies
4 The Proposed Model
4.1 Properties
4.2 Preliminaries
4.3 Model Description
5 Security Analysis and Discussion
6 Conclusions and Future Works
References
GDMA
Chinese Spelling Error Detection and Correction Based on Knowledge Graph
1 Introduction
2 Model
2.1 Knowledge Network
2.2 Detection and Correction Network
2.3 Filter
2.4 Loss
3 Experiments and Evaluation
3.1 Datasets and Baseline
3.2 Evaluation
3.3 Results
4 Related Work
5 Conclusion
References
Construction and Application of Event Logic Graph: A Survey
1 Introduction
2 Event Extraction
2.1 Methods Based on Pattern Matching
2.2 Methods Based on Machine Learning
2.3 Methods Based on Deep Learning
3 Event Relation Extraction
4 Applications
4.1 Detect Hot Events
4.2 Analyze the Event Lineage
4.3 Predict Future Events
5 Summary and Prospect
References
Enhancing Low-Resource Languages Question Answering with Syntactic Graph
1 Introduction
2 Syntactic Information Evolvement in mBERT
3 Related Work
4 Method
4.1 Syntactic Graph
4.2 Syntactic Graph Prediction Task
5 Experiment
5.1 Datasets and Baseline Models
5.2 Setup and Evaluation Metric
5.3 Experiment Results
6 Analysis
6.1 Why Use Parallel Sentence Pairs to Train Syntax Task?
6.2 Why the Syntactic Graph Prediction Task Works?
7 Conclusions
References
Profile Consistency Discrimination
1 Introduction
2 Relate Work
2.1 Natural Language Inference
2.2 Role Consistency
3 Consistency Discrimination
3.1 Problem Definition
3.2 Consistency Discriminator
4 Experiments Setup
4.1 Implementation Details and Evaluations
4.2 Baselines
5 Result
6 Case Study
7 Conclusion and Future Work
References
BDMS
H-V: An Improved Coding Layout Based on Erasure Coded Storage System
1 Introduction
2 Related Work
3 HRS(n,k) - VRS(n',k') Encoding Method
4 Effect Analysis of HRS(n,k) - VRS(n',k')
4.1 Data Transmission Performance Analysis
4.2 Storage Redundancy Analysis
5 Conclusion
References
Astral: An Autoencoder-Based Model for Pedestrian Trajectory Prediction of Variable-Length
1 Introduction
2 Problem Definitions
3 Methodology
3.1 Model Framework
3.2 Autoencoder
3.3 Multi-head Attention
3.4 Online Model
4 Experiments
4.1 Experimental Settings
4.2 Evaluation of Trajectory Prediction
4.3 Ablation Studies
4.4 Case Study
5 Related Work
5.1 Crowd Interaction
5.2 LSTM for Sequence Prediction
6 Conclusion
References
A Survey on Spatiotemporal Data Processing Techniques in Smart Urban Rail
1 Introduction
2 Spatiotemporal Data Processing Technology
3 Applications of Smart Urban Rail
3.1 Intelligent Scheduling
3.2 Intelligent Operation Platform
3.3 Intelligent Perception
3.4 Intelligent Train Control
4 Future Work
5 Conclusion
References
Fast Vehicle Track Counting in Traffic Video
1 Introduction
2 Related Work
3 Preliminary and Problem Statement
4 An Efficient Vehicle Query Counting Method
4.1 Adaptively Choosing Frames for Vehicle Detection
4.2 Vehicle Tracking Based on Location and Simple Appearance Features
4.3 Cascade Track Judgment and Counting
5 Experiments
5.1 Settings
5.2 Metrics
5.3 Experimental Results
5.4 Ablation Experiments
6 Conclusion
References
TSummary: A Traffic Summarization System Using Semantic Words
1 Introduction
2 System Overview
2.1 Feature Extraction
2.2 Preliminary Concepts
2.3 Structure of TSummary
3 Traffic Summarization
3.1 Periodic Feature Detection
3.2 Road Partitioning
3.3 Temporal Merge
4 Experiment
4.1 Experiment Setup
4.2 Evaluation Approach
4.3 Performance Evaluation
5 Related Work
6 Conclusions
References
Attention-Cooperated Reinforcement Learning for Multi-agent Path Planning
1 Introduction
2 Related Work
2.1 Classical Path Planning Methods
2.2 Learning Based Methods
3 Problem Formulation
4 Approach
4.1 State's Structure
4.2 Action Space
4.3 Reward Design
4.4 Model Architecture
5 Experiments
5.1 Experiment Setting
5.2 Training Details
5.3 Metrics
5.4 Baselines
5.5 Results
6 Conclusion
References
Big Data-Driven Stable Task Allocation in Ride-Hailing Services
1 Introduction
2 Related Work
3 Problem Definition
3.1 Preliminaries and Definition
3.2 A Baseline Approach
4 Equilibrium Stable Matching and Global Distance Optimization
4.1 Chain Algorithm
4.2 The Benefit Function for Equilibrium
5 Experimental Study
6 Conclusion
References
Weighted Mean-Field Multi-Agent Reinforcement Learning via Reward Attribution Decomposition
1 Introduction
2 Related Work
3 Preliminaries
3.1 Markov Decision Process and Markov Game
3.2 Mean-Field Reinforcement Learning
4 Algorithm
4.1 Weighted Mean-Field Approximation
4.2 Reward Attribution Decomposition
4.3 Network Architecture
4.4 Overall Optimization Objective
5 Experiments
5.1 Results and Analysis
6 Conclusion
References
BDQM
Evaluating Presto and SparkSQL with TPC-DS
1 Introduction
2 Background
3 Experimental Setup
4 Experimental Result
4.1 TPC-DS on Presto
4.2 TPC-DS on SparkSQL
4.3 Comparison
5 Related Work
6 Conclusion
References
Optimizing the Age of Sensed Information in Cyber-Physical Systems
1 Introduction
2 Problem Definition
2.1 System Model
2.2 Problem Definition
3 Max AoI Optimizing Algorithm in Cyber-Physical Systems
3.1 Determining the AoI of Sensed Data
3.2 Greedy Strategy for Each Available Time
3.3 Scheduling Algorithm
4 Experiment Results
4.1 Experiment Settings
4.2 Baseline
4.3 Performance of SAMMA Algorithm
5 Conclusion
References
Aggregate Query Result Correctness Using Pattern Tables
1 Introduction
2 Related Work
3 Important Concepts
3.1 Reference Table
3.2 Pattern Table and Pattern Tree
3.3 Aggregate Query Execution
4 Pattern Table Algorithm
5 Experimental Setup
6 Results
6.1 Basic Query Execution
6.2 Data Scaling for Query Execution
6.3 Algorithm Execution
7 Conclusion
References
Time Series Data Quality Enhancing Based on Pattern Alignment
1 Introduction
2 Overview
2.1 Preliminaries
2.2 Approach Overview
3 Methodology
3.1 Graph Construction and Pattern Alignment
3.2 Data Quality Evaluation and Enhancing
4 Evaluation
4.1 Experimental Setup
4.2 Case Study
4.3 Adjustments and Experimental Results
5 Conclusion
References
Research on Feature Extraction Method of Data Quality Intelligent Detection
1 Introduction
2 Data Quality Intelligent Detection Method Based on Data Asset List
2.1 The Problem of Power Data
2.2 Anomaly Detection Process
2.3 Word Segmentation Based on Data Asset List
2.4 Keyword Library of Detection Feature
2.5 Keyword Library of Detection Feature
3 Results and Analysis
4 Conclusions
References
MAQTDS
Big Data Resources to Support Research Opportunities on Air Pollution Analysis in India
1 Introduction – Air Quality in India
2 Big Data Research
2.1 Support to Ambient Monitoring Efforts
2.2 Use of Satellite Retrievals
2.3 Use of Meteorological Data
2.4 Use of Google Earth Services
2.5 Use of Google Maps Services
2.6 Use of Open Street Maps (OSM) Database
3 Conclusions
References
Air Quality Data Collection in Hyderabad Using Low-Cost Sensors: Initial Experiences
1 Introduction
2 Materials and Methodology
2.1 Details of the Equipment
2.2 Details of the Locations and Routes
2.3 Data Collection Methodology
3 Results and Observations
3.1 Analysis of Data Collected from Static Monitors
3.2 Analysis of Data Collected from the Selected Routes
4 Implementation Issues
5 Conclusion
References
Visualizing Spatio-temporal Variation of Ambient Air Pollution in Four Small Towns in India
1 Introduction
2 Methodology
2.1 Use of Low-Cost Air Quality Monitors
2.2 Locations of Air Quality Monitors With in Each Study Area
3 Results and Discussion
3.1 Overview of Pollution Data
3.2 Daily Mean PM Variation in the Four Cities
3.3 Monthly Mean PM Variation in the Four Cities
3.4 Diurnal PM Variation in the Four Cities
3.5 PM Isopleths Using Combined Data from Static and Mobile Monitors
4 Discussion and Conclusion
4.1 Summary of Findings
4.2 Discussion
4.3 Future Studies
References
Author Index