ACM TKDD 2024

178 papers
In 2024, ACM Transactions on Knowledge Discovery from Data (TKDD) published 257 papers, with a primary focus on the evolution of graph neural networks (GNNs). Research themes centered on managing complex graph structures, including heterophilic and heterogeneous relational data, and improving model robustness through topological denoising and structural shift bridging. Prolific contributors such as Xinbing Wang, Depeng Jin, and Chuan Chen explored advancements in node representation learning, parameter-agnostic clustering, and the decoupling of ego and neighborhood representations. The volume also featured work on noise robustness and model controllability, utilizing techniques like contrastive representation learning and feature-wise linear modulation. Application-oriented research addressed misinformation detection, multi-label text classification, and the security of click-through rate predictors against fake samples. Additionally, the edition highlighted developments in causality-based fairness for multiple decisions and few-shot node classification. These contributions reflect a broader trend toward balancing multiple objectives in deep learning and enhancing the interpretability of high-dimensional data models.

Top Research Directions

Research directions most represented by ACM TKDD 2024 papers.

Hyperparameter Sensitivity in Complex Machine Learning Frameworks 5
Adapting Homogeneous Graph Learning to Heterogeneous Structures 3
Graph Attention Mechanisms for Adaptive Feature Aggregation 3
Design Space Exploration for Graph Neural Networks 2
Cross-Task Generalization in Graph Neural Networks 2
Overlapping Community Detection in Complex Networks 2
Learning on Graphs with Scarce Labels 2
Overcoming the Homophily Assumption in Graph Neural Networks 2
Over-Smoothing in Graph Neural Networks 2
Pseudo-likelihood Estimation for High-Dimensional Graphical Models 1
Exploiting Density Frontiers in Subgraph Optimization 1
Geometric Artifacts in Low Dimensional Projections 1
Look Ahead Search Space Pruning in Pattern Mining 1
Cross Task Generalization in Time Series Modeling 1
Constraint-Driven Search Space Pruning 1

Top Authors

AuthorPapers
Yong Li6
Chuan Chen4
Depeng Jin4
Xinbing Wang4
Zibin Zheng4
Chenghu Zhou3
Jundong Li3
Kui Yu3
Lele Fu3
Philip S. Yu3
Shichao Zhang3
Weinan Zhang3
Xiao Luo3
Yanjie Fu3
Chong Chen2
Dongjie Wang2
Enhong Chen2
Enneng Yang2
Guibing Guo2
Hanghang Tong2

Impactful Papers

Papers from ACM TKDD 2024 that shaped research clusters.

Influential

Papers

Title
Sequential and Graphical Cross-Domain Recommendations with a Multi-View Hierarchical Transfer Gate
Server-Client Collaborative Distillation for Federated Reinforcement Learning
SPORT: A Subgraph Perspective on Graph Classification with Label Noise
StructCoder: Structure-Aware Transformer for Code Generation
Structural Properties on Scale-Free Tree Network with an Ultra-Large Diameter
Structure-Driven Representation Learning for Deep Clustering
Subspace-Contrastive Multi-View Clustering
Swarm Self-supervised Hypergraph Embedding for Recommendation
TaSPM: Targeted Sequential Pattern Mining
Three-stage Transferable and Generative Crowdsourced Comment Integration Framework Based on Zero- and Few-shot Learning with Domain Distribution Alignment
TOMGPT: Reliable Text-Only Training Approach for Cost-Effective Multi-modal Large Language Model
Totally-ordered Sequential Rules for Utility Maximization
Toward Cross-Lingual Social Event Detection with Hybrid Knowledge Distillation
Toward Few-Label Vertical Federated Learning
Towards Differential Privacy in Sequential Recommendation: A Noisy Graph Neural Network Approach
Towards Domain-Aware Stable Meta Learning for Out-of-Distribution Generalization
Towards Faster Deep Graph Clustering via Efficient Graph Auto-Encoder
Traceable Group-Wise Self-Optimizing Feature Transformation Learning: A Dual Optimization Perspective
Trajectory-User Linking via Hierarchical Spatio-Temporal Attention Networks
Transfer Learning across Graph Convolutional Networks: Methods, Theory, and Applications
TTS-Norm: Forecasting Tensor Time Series via Multi-Way Normalization
Understanding and Guiding Weakly Supervised Entity Alignment with Potential Isomorphism Propagation
Understanding Any Time Series Classifier with a Subsequence-based Explainer
Urban Knowledge Graph Aided Mobile User Profiling
Utility-Oriented Reranking with Counterfactual Context
VAE*: A Novel Variational Autoencoder via Revisiting Positive and Negative Samples for Top- N Recommendation
VAE*: A Novel Variational Autoencoder via Revisiting Positive and Negative Samples for Top-Recommendation
Variate Associated Domain Adaptation for Unsupervised Multivariate Time Series Anomaly Detection

« prev Page 4/4