
Database Systems for Advanced Applications
Description
This six-volume set LNCS 15986-15991 constitutes the proceedings of the 30th International Conference on Database Systems for Advanced Applications, DASFAA 2025, held in Singapore, during May 26-29, 2025.
The 136 full papers presented in this book together with 89 short papers were carefully reviewed and selected from 731 submissions.They cover topics such as
Part I-
Machine Learning and Text.
Part II-
Emerging Application; NLP and Spatial-Temporal.
Part III- Graph; Knowledge Graph.
Part IV- Database core; Fedrated Learning; Social; Steams & Time Series.
Part V- Recommendation and Security & Privacy.
Part VI- Language Model; Industry Papers and Demo Papers.
More details
Content
Database Core. - Well-designed Query Optimization Based on Pattern Tree.- Separating Frozen Pages via Learning-based Recognition with ZNS SSD for Write Amplification Reduction in Database.- BlindChain: Keeping Query Privacy in Blockchain Out of Sight.- OmniQO: An Adaptive Framework for Integrating ML and Traditional Query Optimizers.- LASE: A Learned Spatial Index for Dynamic Workloads.- HiCHT: High-performance Compact Hash Table.- WorthyPar: A Workload-Aware Data Hybrid Partitioning Advisor with Deep Reinforcement Learning.- GAS-DBSCAN: A Grid-based Adaptive Sampling Method for DBSCAN Clustering under Skewed Data Distribution.- Learning Distance-Aware Space Partitions for Approximate Nearest Neighbor Search.- Perspective-based Multi-task Learning for Outlier Interpretation.- SELVA: A Reliable and Fast Selectivity Estimation Method for Query Plan Optimization in Video Analytics.- Transcending Conventional Binary Labels: Revamping Knowledge Tracing with VAE-Generated Image Representation.- Efficient Computation of k Representative Regret Minimization G-Skyline Groups.- MoEPlan: A Lazy Learned Query-Selection Optimizer via Mixture of Optimizer Experts.- Time-Optimal Route Planning for Non-Linear Recharging Electric Vehicles on Road Networks. - Federated Learning. - RAP: Random Projection is What You Need for Vertical Federated Learning.- VF-FD: Feature Deduplication for Vertical Federated Learning.- VHFed: A Two-Tier Vertical and Horizontal Federated Learning Framework for Enhanced Model Performance.- Heterogeneous FL via active-passive collaboration.- Information-agnostic Model Poisoning Attacks against Byzantine-robust Federated Learning. - Social. - A Diffusion-based Triple Embedding Model for User Identity Linkage across Social Networks.- Key Users Identification-based Heterogeneous Hypergraph for Group Recommendation.- PRIM: Encoding Propagation Probability and Role-Aware Representation for Influence Maximization.- Clustering-Guided Dynamic Social Network Graph Partitioning.- Retrieval-Based Multimodal Data Augmentation for Multimodal Information Extraction in Social Media.- KMMN: Knowledge Enhanced Multimodal Multi-grained Network for Fake News Detection. - Steams & Time Series. - Memory-Augmented Short Time Series Forecasting.- Dynamic Group Nearest Neighbor Group Query over Streaming Data.- Compress Time Series with Smaller Error Tolerances.- UniMixer: Unified Patch-Wise and Global Inter-Series Dependency Modeling for Multivariate Time Series Forecasting.- Dynamic Multiple Continuous Top-k Queries Over Streaming Data.- CausalScaler: A Causality-Driven Autoscaling Framework for the Cloud.- A Benchmark Dataset and Instruction Fine-Tuning Methods for Metaphorical Comprehension and Explanation.- MPPG: Pluggable Multi-Periodic Pattern-Guided Approach for Multivariate Time Series Anomaly Detection.