ICDE 2021

228 papers
The 2021 IEEE International Conference on Data Engineering (ICDE) featured 351 papers addressing a wide range of data management and analysis challenges. Prominent research themes included database cracking for on-the-fly data reorganization and the integration of exogenous context into spatio-temporal forecasting. A substantial portion of the program focused on the intersection of machine learning and database systems, specifically regarding learned cardinality estimation for query optimization, standardized evaluation protocols, and unified benchmarking for neural architectures. Technical contributions also explored storage systems and query processing, including SSD controller programming for variable-size pages and the development of specialized key-value stores like WipDB and LogStore. Other research areas included knowledge graph database systems, secure dynamic skyline queries, and popularity forecasting based on early interaction dynamics. Prolific authors contributing to this edition included Xuemin Lin, Xiaofang Zhou, and Jianliang Xu. The year's work demonstrated a continued interest in optimizing system performance through automated synthesis from scientific literature and bandit algorithms for black-box system optimization.

Top Research Directions

Research directions most represented by ICDE 2021 papers.

Query Driven Incremental Data Reorganization 2
Hyperparameter Sensitivity in Complex Machine Learning Frameworks 2
Learning in Non-Stationary Data Streams 2
Anomaly Detection via Data Compression 1
Detecting Distribution Shifts via Reconstruction Error 1
Structural Diversity in Information Diffusion 1
Mitigating Cardinality Estimation Errors in Database Query Optimizers 1
Accelerating Queries through Strategic Precomputation and View Rewriting 1
Geometric Duality Transformations for Spatial Optimization 1
Bandit Guided Strategy Selection in High Dimensional Spaces 1
Modeling Exogenous Influence in Social and Information Networks 1
Optimizing Node Layouts in Hierarchical Indexes 1
The Search Verification Gap 1
Enforcing Global One-to-One Constraints in Alignment Tasks 1
Design Space Exploration for Graph Neural Networks 1

Top Authors

AuthorPapers
Xuemin Lin10
Jianliang Xu7
Lei Chen7
Xiaofang Zhou7
Lu Qin6
Wenjie Zhang6
Ye Yuan5
Ying Zhang5
Yunjun Gao5
Bolong Zheng4
Christian S. Jensen4
Guoren Wang4
Hongzhi Yin4
Lu Chen4
Yuchen Li4
Zhao Li4
Aoying Zhou3
Beng Chin Ooi3
Byron Choi3
Cheng Long3

Impactful Papers

Papers from ICDE 2021 that shaped research clusters.

Influential

Papers

Title
Group-Buying Recommendation for Social E-Commerce
G-TADOC: Enabling Efficient GPU-Based Text Analytics without Decompression
Hash Adaptive Bloom Filter
Hate is the New Infodemic: A Topic-aware Modeling of Hate Speech Diffusion on Twitter
Heterogeneous Information Assisted Bandit Learning: Theory and Application
High-Dimensional Similarity Search for Scalable Data Science
High-Performance Smart Contracts Concurrent Execution for Permissioned Blockchain Using SGX
HST+: An Efficient Index for Embedding Arbitrary Metric Spaces
Hypercore Maintenance in Dynamic Hypergraphs
Improving Constrained Search Results By Data Melioration
Improving Conversational Recommender System by Pretraining Billion-scale Knowledge Graph
Index-based Solutions for Efficient Density Peak Clustering (Extended Abstract)
Influence Maximization Based on Dynamic Personal Perception in Knowledge Graph
INFOSHIELD: Generalizable Information-Theoretic Human-Trafficking Detection
IntelliTag: An Intelligent Cloud Customer Service System Based on Tag Recommendation
Joint Index, Sorting, and Compression Optimization for Memory-Efficient Spatio-Temporal Data Management
Jointly Attacking Graph Neural Network and its Explanations
KDDLog: Performance and Scalability in Knowledge Discovery by Declarative Queries with Aggregates
Knowledge-Aware Group Representation Learning for Group Recommendation
Knowledge-Based Dynamic Systems Modeling: A Case Study on Modeling River Water Quality
Large-scale Fake Click Detection for E-commerce Recommendation Systems
Learnings from a Retail Recommendation System on Billions of Interactions at bol.com
Learning to Optimize Industry-Scale Dynamic Pickup and Delivery Problems
Less is More: De-amplifying I/Os for Key-value Stores with a Log-assisted LSM-tree
Leveraging Currency for Repairing Inconsistent and Incomplete Data (Extended Abstract)
Leveraging Meta-path Contexts for Classification in Heterogeneous Information Networks
Leveraging Temporal and Topological Selectivities in Temporal-clique Subgraph Query Processing
LHist: Towards Learning Multi-dimensional Histogram for Massive Spatial Data
LineageBA: A Fast, Exact and Scalable Graph Generation for the Barabási-Albert Model
LogLog Filter: Filtering Cold Items within a Large Range over High Speed Data Streams
LogStore: A Workload-aware, Adaptable Key-Value Store on Hybrid Storage Systems (Extended abstract)
LShape Partitioning: Parallel Skyline Query Processing using MapReduce (Extended Abstract)
Manipulating Black-Box Networks for Centrality Promotion
MaxiZone: Maximizing Influence Zone over Geo-Textual Data (Extended Abstract)
Meepo: Sharded Consortium Blockchain
Memorize, Factorize, or be Naive: Learning Optimal Feature Interaction Methods for CTR Prediction
Memory-Efficient Database Fragment Allocation for Robust Load Balancing when Nodes Fail
MLCask: Efficient Management of Component Evolution in Collaborative Data Analytics Pipelines
Modeling Citywide Crowd Flows using Attentive Convolutional LSTM
Multi-attributed Community Search in Road-social Networks
Multi-Behavior Enhanced Recommendation with Cross-Interaction Collaborative Relation Modeling
Multi-Facet Recommender Networks with Spherical Optimization
Near-Optimal Fixed-Route Scheduling for Crowdsourced Transit System
NestGPU: Nested Query Processing on GPU
NewsLink: Empowering Intuitive News Search with Knowledge Graphs
Noah: Neural-optimized A* Search Algorithm for Graph Edit Distance Computation
Node2LV: Squared Lorentzian Representations for Node Proximity
Odess: Speeding up Resemblance Detection for Redundancy Elimination by Fast Content-Defined Sampling
On Disambiguating Authors: Collaboration Network Reconstruction in a Bottom-up Manner
On Efficient and Scalable Time-Continuous Spatial Crowdsourcing

« prev Page 3/5 next »