Network Science Graph Analytics

Temporal Dynamics in Evolving Networks

Studies how network structure and properties change over time, including birth, growth, and decay of nodes and edges in dynamic systems.

RESEARCH GAP FRONTIERS

Community Detection and Clustering Algorithms

Develops and analyzes methods for identifying densely connected subgroups within networks using modularity optimization and spectral approaches.

RESEARCH GAP FRONTIERS

Graph Neural Networks for Node Classification

Applies deep learning on graph-structured data to predict node labels and attributes using convolutional and attention-based architectures.

RESEARCH GAP FRONTIERS

Link Prediction in Knowledge Graphs

Develops models to predict missing relationships in semantic and knowledge graphs using embedding methods and graph completion techniques.

RESEARCH GAP FRONTIERS

Centrality Measures and Node Importance

Investigates various centrality metrics beyond degree and betweenness to identify influential nodes in complex network structures.

RESEARCH GAP FRONTIERS

Scale-Free Network Properties and Analysis

Examines power-law degree distributions and self-similar properties in real-world networks like the internet and biological systems.

RESEARCH GAP FRONTIERS

Biological Network Reconstruction Methods

Develops techniques to infer protein-protein interactions, gene regulatory networks, and metabolic pathways from high-dimensional biological data.

RESEARCH GAP FRONTIERS

Social Network Analysis and Influence Propagation

Studies information diffusion, influence maximization, and cascade models in social networks with applications to viral marketing and epidemiology.

Heterogeneous Network Embedding Methods

Develops representation learning techniques for networks with multiple types of nodes and edges using meta-path and attention mechanisms.

Graph Convolutional Networks and Extensions

Advances graph convolutional architectures including spectral methods, spatial convolutions, and variants for diverse network learning tasks.

Network Motif Discovery and Analysis

Identifies and characterizes recurring subgraph patterns that reveal functional modules and design principles in biological and engineered networks.

Multilayer and Multiplex Network Analysis

Analyzes interdependent networks with multiple relationship types using tensor methods and layer-aware centrality measures.

Graph Isomorphism and Subgraph Matching

Develops algorithms for comparing graph structures, detecting isomorphism classes, and matching subgraphs in large-scale networks.

Network Robustness and Resilience Analysis

Studies vulnerability to node and edge failures, cascade failures, and strategies for improving network fault tolerance and recovery.

Signed Network Analysis and Balance Theory

Analyzes networks with positive and negative edges to understand structural balance, conflict dynamics, and polarization phenomena.

Graph Attention Networks and Transformers

Develops attention mechanisms for graphs enabling selective aggregation of neighborhood information in transformer-based architectures.

Random Walk Algorithms and Applications

Applies random walk theory including DeepWalk, Node2Vec, and PageRank for node embedding and ranking in massive graphs.

Supply Chain Network Optimization

Models and optimizes complex supply chain networks for resilience, efficiency, and sustainability using graph-based methods.

Brain Connectome Analysis and Neuroscience Networks

Analyzes structural and functional brain networks to understand connectivity patterns, neural synchronization, and cognitive mechanisms.

Anomaly Detection in Dynamic Graphs

Develops methods to identify unusual patterns, fraudulent edges, and abnormal behaviors in evolving network structures.

Graph Generation and Generative Models

Creates generative models including VAEs and GANs for producing synthetic graphs with realistic properties and structure.

Explainability in Graph Neural Networks

Develops interpretability methods to understand GNN predictions through node importance, edge attribution, and graph visualization techniques.

Citation Network Analysis and Scientometrics

Analyzes academic collaboration and citation networks to identify research trends, influential authors, and interdisciplinary connections.

Network Comparison and Similarity Metrics

Develops distance measures and similarity metrics for comparing entire networks including graph kernels and Weisfeiler-Lehman methods.

Financial Network Systemic Risk Analysis

Analyzes financial institution networks to assess systemic risk, contagion effects, and interconnectedness in banking systems.

Transportation Network Design and Routing

Optimizes vehicle routing, traffic flow, and public transportation networks using graph algorithms and network design techniques.

Graph Pooling and Readout Mechanisms

Develops hierarchical pooling strategies and graph-level readout functions for end-to-end graph classification and regression.

Spectral Graph Theory Applications

Applies eigenvalue and eigenvector properties of adjacency matrices to analyze network structure and design efficient algorithms.

Recommender Systems using Graph Methods

Builds recommendation engines using user-item bipartite graphs, collaborative filtering, and graph-based collaborative approaches.

Ecological Network and Food Web Analysis

Analyzes species interaction networks and food webs to understand ecosystem stability, biodiversity, and trophic dynamics.

Named Entity Recognition in Knowledge Bases

Extracts and disambiguates entities in text to construct and enrich structured knowledge graphs using NLP and graph methods.

Graph Contrastive Learning Frameworks

Develops self-supervised learning methods using contrastive objectives to learn graph representations without labeled data.

Message Passing Neural Networks

Formalizes node updates through iterative message passing between neighbors to create flexible and powerful graph learning models.

Hypergraph Analysis and Higher-Order Networks

Extends graph analysis to hypergraphs with higher-order interactions, group relationships, and simplicial complex structures.

Coevolutionary Network Dynamics

Studies coupled evolution of network topology and node states including coupled oscillators and adaptive networks.

Entity Alignment in Heterogeneous Knowledge Graphs

Develops techniques to match equivalent entities across multiple knowledge graphs using cross-lingual and cross-domain alignment methods.

Graph Sampling and Subgraph Methods

Creates efficient sampling strategies for processing large-scale graphs including neighborhood sampling and variance reduction techniques.

Adversarial Robustness of Graph Networks

Studies adversarial attacks on graph neural networks and develops defense mechanisms against perturbations in graph structure and features.

Machine Learning Compilation and Optimization

Applies graph-based program representations and compute graph optimization for efficient neural network compilation and execution.

Graph Matching and Graph Edit Distance

Develops algorithms for optimal graph matching, edit distance computation, and alignment with applications to graph similarity.

Heterogeneous Information Network Mining

Extracts patterns and knowledge from complex networks with multiple node and edge types using meta-path mining and clustering.

Graph Sparsification and Simplification

Creates sparse graph approximations that preserve key structural properties while reducing computational complexity and memory requirements.

Federated Learning on Distributed Graphs

Develops decentralized graph learning algorithms for privacy-preserving training on networks distributed across multiple parties.

Knowledge Graph Completion and Reasoning

Advances methods for inferring missing facts and performing logical reasoning in knowledge graphs using embeddings and rule mining.

Graph Autoencoder Architectures

Develops encoder-decoder frameworks for unsupervised graph representation learning and graph reconstruction tasks.

Network Flow and Bottleneck Analysis

Analyzes information and material flow through networks to identify bottlenecks, criticality, and efficiency of transport processes.

Graph Kernels for Classification

Develops kernel functions on graphs enabling support vector machines and kernel methods for whole-graph classification tasks.

Attributed Network Clustering and Classification

Combines node attributes and network topology for improved clustering and classification in networks with rich node feature information.

Quantum Graph Algorithms

Develops quantum computing algorithms for graph problems including quantum walks, search, and optimization on quantum computers.

Graph Time Series Forecasting

Combines temporal dynamics with graph structure to forecast node attributes and network evolution using spatio-temporal models.

Dynamic Graph Neural Network Architectures

Research on designing neural network models that efficiently process and adapt to graphs that change over time with evolving node and edge attributes.

Graph Fairness and Bias Mitigation

Investigation of fairness principles in graph learning algorithms and development of techniques to detect and eliminate bias in graph-based predictions.

Spatial-Temporal Graph Convolution Networks

Development of convolutional architectures that simultaneously model spatial graph structure and temporal dependencies for time-series prediction on networks.

Graph Deep Learning for Drug Discovery

Application of graph neural networks to molecular structure representation and prediction of drug efficacy and toxicity properties.

Continuous-Time Graph Representation Learning

Development of embedding methods for graphs with continuous temporal information enabling smooth interpolation and extrapolation of node states.

Privacy-Preserving Graph Analytics

Research on differential privacy, federated learning, and cryptographic techniques applied to protect sensitive information in network analysis.

Graph Neural Networks for Combinatorial Optimization

Application of GNNs to solve NP-hard optimization problems such as traveling salesman problem and maximum cut through learned heuristics.

Causal Inference on Network Data

Development of methods to identify causal relationships and treatment effects in networked data accounting for network interference and spillover effects.

Graph Representation Learning at Scale

Techniques for efficiently computing node and graph embeddings on billion-scale networks using distributed computing and memory-efficient algorithms.

Interpretable Graph Classification Models

Research on developing graph classification methods that provide human-interpretable explanations for predictions through subgraph importance and attention mechanisms.

Cross-Domain Knowledge Graph Transfer Learning

Methods for transferring learned representations and models from knowledge graphs in one domain to improve learning in structurally different domains.

Dynamic Community Detection in Streaming Networks

Algorithms for identifying and tracking evolving communities in networks where edges arrive continuously with minimal memory and computational overhead.

Graph Regularization for Semi-Supervised Learning

Techniques using graph structure to regularize learning objectives and improve generalization when labeled training data is scarce.

Relationship Extraction from Unstructured Text

Methods for extracting entity relationships from documents and constructing knowledge graphs using natural language processing and graph neural networks.

Network Inference from Observational Data

Algorithms to infer hidden network structures from indirect observations such as correlation matrices or activity patterns without direct edge measurements.

Attention Mechanisms for Heterogeneous Graphs

Design of attention-based neural architectures that weight different node and edge types differently for improved prediction in multi-type networks.

Graph Neural Networks for Traffic Prediction

Application of GNNs to predict traffic flow and congestion in transportation networks using road network topology and historical traffic patterns.

Meta-Learning on Graph Datasets

Research on few-shot and zero-shot learning techniques that enable graph neural networks to quickly adapt to new tasks with minimal examples.

Graph Pooling via Differentiable Clustering

Development of learnable graph coarsening methods using differentiable clustering to create hierarchical graph representations.

Protein Interaction Network Prediction

Methods for predicting protein-protein interactions and discovering functional modules in biological networks using graph learning approaches.

Urban Computing with Spatial Networks

Analysis of city-scale networks integrating transportation, social, and infrastructure networks to solve urban planning and smart city challenges.

Graph Structure Learning and Optimization

Methods that jointly learn both the graph structure and node representations, optimizing connectivity to improve downstream prediction tasks.

Scalable Influence Maximization Algorithms

Efficient algorithms for identifying influential nodes and seed sets that maximize information diffusion in large-scale social networks.

Tensor Methods for Network Analysis

Application of tensor decomposition and higher-order tensor techniques to analyze multi-dimensional network data and discover latent factors.

Graph Neural Network Pruning and Compression

Techniques for reducing the computational complexity and model size of graph neural networks while maintaining predictive performance.

Disinformation Detection in Information Networks

Graph-based methods for detecting fake news and misinformation by analyzing information spread patterns and source credibility networks.

Inductive Graph Representation Learning

Development of methods that can generate embeddings for unseen nodes and graphs without retraining, enabling deployment in dynamic environments.

Hypergraph Neural Networks and Learning

Neural network architectures specifically designed for hypergraphs where edges can connect multiple nodes, capturing higher-order interactions.

Network Reconstruction from Incomplete Data

Methods for accurately reconstructing network topology and estimating missing edges when observational data is sparse or contains measurement errors.

Graph-Based Sentiment Analysis Networks

Techniques for performing sentiment analysis by constructing aspect and opinion networks and leveraging graph structure for improved classification.

Anomaly Detection in Knowledge Graphs

Methods for identifying incorrect, fraudulent, or inconsistent facts and relationships in knowledge graphs using graph-based anomaly detection.

Graph Inverse Problems and Network Deconvolution

Research on recovering underlying network structure from aggregated or filtered observations using inverse problem formulations.

Reinforcement Learning on Graph Structures

Development of reinforcement learning agents that operate on graph-structured environments for optimization and decision-making tasks.

Graph Benchmark Datasets and Evaluation

Creation of standardized benchmark datasets and evaluation protocols for fair comparison of graph learning algorithms across diverse domains.

Adversarial Attacks on Graph Classifiers

Study of adversarial perturbations on graph structure and features that fool neural networks, and defenses against such attacks.

Ethereum Blockchain Network Analysis

Graph-based analysis of blockchain networks to detect fraudulent transactions, money laundering patterns, and identify key actors.

Temporal Point Processes on Networks

Probabilistic models for event sequences on networks that capture both network structure and temporal dependencies in event occurrences.

Graph Edit Distance and Metric Learning

Development of efficient algorithms for computing graph similarity metrics and learning distance functions for graph-structured data.

Corporate Ownership Network Analysis

Analysis of company ownership and control networks to identify beneficial owners, detect conflicts of interest, and assess economic concentration.

Collaborative Filtering with Graph Methods

Integration of graph neural networks with user-item bipartite networks to improve recommendation accuracy and address cold-start problems.

Neural Architecture Search for Graphs

Automated methods for designing optimal graph neural network architectures through architecture search for different application domains.

Epidemic Spreading Models on Networks

Mathematical and computational models of disease or information spread through networks with applications to pandemic prediction and control.

Heterogeneous Temporal Network Analysis

Methods for analyzing networks with multiple node and edge types that evolve over time, capturing complex interaction patterns.

Protein Function Prediction Networks

Graph-based approaches for predicting unknown protein functions by leveraging biological network information and functional annotations.

Signed Knowledge Graph Embeddings

Embedding techniques for knowledge graphs containing both positive and negative relationships, preserving relational semantics.

Graph Deep Learning for Materials Science

Application of graph neural networks to predict material properties from atomic structures and discover new materials with desired characteristics.

Multi-Hop Reasoning in Knowledge Graphs

Methods for performing logical reasoning over knowledge graphs by finding and aggregating paths of multiple hops between entities.

Graph Clustering and Partitioning Algorithms

Advanced techniques for partitioning large graphs into balanced clusters for parallel processing and understanding network modularity.

Social Recommendation with Trust Networks

Integration of explicit trust relationships alongside user-item networks to provide more reliable and trustworthy recommendations.

Graph Representation Learning via Autoregressive Models

Investigates autoregressive approaches for learning node and edge representations in large-scale graphs with sequential dependencies and temporal ordering constraints.

Causal Inference in Observational Network Data

Develops causal discovery and inference methodologies specifically designed for identifying treatment effects and causal relationships within complex network structures.

Fairness and Bias in Graph Machine Learning

Examines algorithmic fairness, bias mitigation, and equitable treatment in graph neural networks and network-based machine learning systems.

Graph Privacy and Differential Privacy Mechanisms

Develops privacy-preserving techniques and differential privacy frameworks for protecting sensitive information in graph data and network analysis.

Explainable AI for Graph Neural Networks

Creates interpretable and transparent explanation methods for understanding decision-making processes in complex graph neural network models.

Network Topology Optimization using Reinforcement Learning

Applies reinforcement learning techniques to discover optimal network structures and topologies for specific computational or operational objectives.

Cross-Domain Graph Transfer Learning

Investigates transfer learning approaches that enable knowledge learned from source graphs to be effectively applied to target network domains.

Graph Contrastive Learning with Self-Supervision

Studies self-supervised contrastive learning methods that leverage graph augmentations and invariance properties without requiring labeled network data.

Meta-Learning for Few-Shot Graph Tasks

Develops meta-learning frameworks enabling rapid adaptation to graph analysis tasks with minimal labeled examples through algorithm learning.

Graph Signal Processing and Filtering

Applies signal processing theory to graph-structured data including spectral filtering, graph wavelet transforms, and frequency domain analysis.

Influence Maximization in Social Networks

Develops algorithms for identifying optimal seed nodes and strategies to maximize information diffusion and influence spread in social network systems.

Protein Interaction Network Prediction Methods

Creates machine learning approaches for predicting protein-protein interactions, complex formation, and functional relationships in biological networks.

Graph Variational Autoencoders and Generative Models

Designs variational and probabilistic generative models for creating realistic synthetic graphs and sampling from learned graph distributions.

Network Motif Enrichment and Statistical Significance

Develops statistical testing frameworks and computational methods for identifying over-represented network motifs and assessing their biological significance.

Temporal Point Processes on Networks

Models event sequences and temporal dynamics in networks using point process theory to capture timing patterns and inter-event dependencies.

Disease Spread Modeling in Contact Networks

Develops epidemiological models and simulation frameworks for understanding disease propagation through contact networks and predicting outbreak dynamics.

Graph Clustering with Uncertain and Noisy Data

Addresses clustering challenges in networks with imperfect information, measurement errors, and probabilistic edge weights through robust methodologies.

Attributed Heterogeneous Network Embedding

Develops embedding techniques that jointly model diverse node types, attributes, and edge relationships in complex heterogeneous network systems.

Graph Neural Networks for Drug Discovery

Applies graph neural networks to molecular property prediction, drug-target interaction discovery, and computational chemistry applications.

Network Reconstruction from Partial and Indirect Data

Develops inference algorithms for reconstructing complete network structures from incomplete observations, aggregated data, or indirect measurements.

Graph Active Learning and Uncertainty Sampling

Designs active learning strategies that intelligently select informative nodes or edges for labeling to improve graph model training efficiency.

Continuous-Time Dynamic Network Embedding

Creates embedding methods that capture continuous temporal dynamics in networks using temporal point processes and neural ordinary differential equations.

Graph Clustering Coefficient and Transitivity Analysis

Analyzes local and global clustering properties, transitivity patterns, and their relationship to network formation and functional organization.

Scalable Graph Processing for Billion-Scale Networks

Develops distributed and parallel algorithms for processing and analyzing extremely large networks with billions of nodes and edges efficiently.

Graph Attention and Explainability via Saliency Maps

Creates visual explanation methods using attention maps and saliency techniques to interpret which graph components influence model predictions.

Community Structure Evolution in Temporal Networks

Tracks the emergence, merging, splitting, and dissolution of communities over time in dynamic networks using evolutionary analysis methods.

Graph Regularization Techniques for Semi-Supervised Learning

Develops regularization methods leveraging graph structure and topology to improve semi-supervised learning with limited labeled network data.

Heterogeneous Temporal Knowledge Graph Completion

Creates methods for completing missing facts in temporal knowledge graphs containing multiple entity and relation types with time-dependent semantics.

Graph Embedding Quality Evaluation Metrics

Proposes comprehensive evaluation frameworks and metrics for assessing the quality and effectiveness of learned graph embeddings.

Equitable Resource Allocation in Network Systems

Develops optimization algorithms for fair and efficient allocation of resources across network nodes considering fairness and efficiency objectives.

Graph Neural Networks for Combinatorial Optimization

Applies graph neural networks to solve NP-hard combinatorial optimization problems such as traveling salesman, routing, and scheduling problems.

Multilingual Knowledge Graph Alignment and Integration

Develops methods for aligning and integrating knowledge graphs across languages and knowledge bases to enable cross-lingual reasoning and retrieval.

Graph Self-Loops and Self-Attention Mechanisms

Investigates the role of self-loops and self-attention in graph neural networks for capturing node-level features and improving model expressivity.

Synthetic Benchmark Graphs for Algorithm Evaluation

Creates synthetic graph generation frameworks with controlled properties for rigorous evaluation and comparison of graph analysis algorithms.

Graph Classification using Geometric Deep Learning

Develops geometric deep learning approaches for classifying entire graphs and predicting graph-level properties using invariant and equivariant representations.

Network Influence Estimation using Causal Models

Applies causal inference techniques to estimate causal influence and direct effects in networks while controlling for confounding factors.

Probabilistic Graphical Models and Bayesian Networks

Studies probabilistic graphical models, Bayesian networks, and factor graphs for uncertainty quantification and probabilistic reasoning in networks.

Graph Neural Networks for Quantum Chemistry

Applies graph neural networks to quantum chemistry problems including molecular orbital prediction and quantum property calculation.

Interpretable Graph-Based Recommender Systems

Develops transparent and interpretable recommendation algorithms using graph structure to provide explainable recommendations to users.

Graph Sparsification via Spectral Methods

Creates spectral graph sparsification techniques that preserve important properties while reducing edge count for efficient computation.

Network-Based Anomaly Detection using Machine Learning

Develops machine learning methods for detecting anomalies, intrusions, and fraudulent patterns in networked systems and graph data.

Graph Matching with Approximate and Heuristic Methods

Designs approximate and heuristic algorithms for solving graph matching and subgraph isomorphism problems in polynomial time.

Persistent Homology and Topological Data Analysis

Applies persistent homology and topological data analysis methods to extract meaningful topological features from graph structures.

Graph Coarsening and Multilevel Graph Hierarchies

Develops hierarchical graph coarsening techniques that progressively aggregate nodes and edges while preserving essential network properties.

Geospatial Network Analysis and Spatial Graphs

Analyzes networks embedded in geographic space considering spatial constraints, distance metrics, and location-dependent network properties.

Graph Neural Networks for Time Series Prediction

Develops graph neural network architectures for multivariate time series forecasting using graph structures to capture variable relationships.

Structural Controllability and Observability in Networks

Studies structural properties determining network controllability and observability for understanding dynamic system behavior over networks.

Graph-Based Clustering for Genomic Data Analysis

Applies graph-based clustering techniques to genomic sequence analysis, gene expression networks, and biological pathway discovery.

Subgraph Mining and Frequent Pattern Discovery

Develops efficient algorithms for discovering frequent subgraph patterns and mining common structural motifs in large graph databases.

Fairness and Bias in Graph Learning

Investigates algorithmic fairness, demographic parity, and bias mitigation techniques in graph neural networks and network analysis algorithms.

Dynamic Influence Maximization Strategies

Develops algorithms for identifying optimal seed sets in time-evolving networks to maximize information spread and cascade effects.

Graph Diffusion Kernels and Smoothness

Studies diffusion processes on graphs and kernel methods that leverage graph structure for semi-supervised learning and node representation.

Interactive Graph Visualization and Exploration

Creates innovative visualization methods and interactive tools for exploring large-scale networks and uncovering hidden structural patterns.

Symbolic Graph Reasoning and Logic

Combines symbolic reasoning and first-order logic with graph representations for interpretable knowledge representation and inference.

Graph Neural Network Scalability Solutions

Addresses computational bottlenecks through efficient mini-batch training, gradient checkpointing, and memory-optimized architectures for billion-scale graphs.

Causal Inference in Observational Networks

Applies causal discovery and causal inference methods to identify treatment effects and causal relationships within network-structured data.

Signed and Weighted Network Dynamics

Analyzes complex networks with signed edges and weighted interactions to model cooperation, conflict, and influence asymmetries.

Incomplete and Partially Observed Graphs

Develops methods for network analysis and inference from incomplete, censored, or partially observed network data with missing edges and nodes.

Graph Augmentation for Semi-Supervised Learning

Designs graph perturbation and augmentation strategies including node dropping, edge masking, and subgraph sampling for robust representation learning.

Heterogeneous Temporal Knowledge Graphs

Models and reasons over knowledge graphs with multiple entity and relation types evolving across time with temporal validity constraints.

Network Reconstruction from Noisy Signals

Recovers true network structure from noisy, incomplete, or indirect observations using inverse problems and statistical inference frameworks.

Graph Structure Learning and Discovery

Infers underlying graph structure from data features and observations using differentiable structure learning and adaptive connectivity mechanisms.

Topological Data Analysis for Networks

Applies persistent homology, simplicial complexes, and topological methods to identify robust structural features and higher-order patterns in networks.

Privacy-Preserving Graph Machine Learning

Develops differential privacy, federated learning, and cryptographic techniques to protect sensitive information in graph analytical tasks.

Synthetic Graph Generation Benchmarks

Creates standardized benchmarks and evaluation metrics for assessing quality and diversity of synthetically generated networks.

Recommender Systems for Knowledge Graphs

Designs recommendation algorithms leveraging knowledge graph structure, relations, and semantic information for personalized entity suggestions.

Multi-Relational Graph Embedding Methods

Develops embedding techniques for graphs with multiple edge types and relations, preserving semantic relationships in vector space.

Graph Neural Network Uncertainty Quantification

Integrates Bayesian methods, ensemble techniques, and Monte Carlo dropout to quantify prediction uncertainty in graph neural networks.

Epidemic Spreading in Contact Networks

Models disease transmission, virus spread, and information cascades in contact networks using compartmental and agent-based models.

Clustering Coefficient and Transitivity Analysis

Studies local and global clustering patterns, triangular motifs, and transitivity properties to characterize network cohesion and community structure.

Graph Representation Learning for Molecules

Develops graph neural networks for molecular structure representation enabling molecular property prediction and drug discovery applications.

Network Alignment and Matching Methods

Develops algorithms for aligning and matching nodes across multiple networks with partial correspondence and structural similarity.

Graph-Based Anomaly Detection Systems

Designs anomaly detection systems that identify unusual patterns, outlier nodes, and abnormal subgraphs in complex networks.

Recurrent Graph Neural Networks Architecture

Develops recurrent and gated mechanisms within graph neural networks for capturing temporal dependencies and long-range interactions.

Stochastic Block Model Extensions

Extends stochastic block models with overlapping communities, hierarchical structures, and degree-corrected variants for realistic network modeling.

Protein Interaction Network Analysis

Analyzes protein-protein interaction networks to predict novel interactions, identify functional modules, and understand biological processes.

Graph Coarsening and Hierarchical Methods

Develops hierarchical graph coarsening, multi-scale analysis, and graph contraction methods for computational efficiency and multi-resolution understanding.

Recommender Diversity in Graph-Based Systems

Designs graph-based recommendation algorithms that balance accuracy with diversity, novelty, and coverage of items or entities.

Neural Architecture Search for Graphs

Automates discovery of optimal graph neural network architectures through reinforcement learning and evolutionary search strategies.

Relation Extraction from Text Networks

Extracts structured relations and entity relationships from unstructured text using graph neural networks and syntactic tree structures.

Graph-Based Active Learning Strategies

Develops active learning methods that query informative nodes in networks to minimize labeling costs while maximizing model performance.

Deepfake Detection in Social Graphs

Detects fake accounts, bots, and fraudulent behavior patterns in social networks using graph-based anomaly detection and network analysis.

Graph Embedding Evaluation Methodologies

Develops comprehensive evaluation frameworks and benchmark datasets for assessing quality of graph embeddings across diverse tasks.

Heterogeneous Graph Attention Mechanisms

Designs attention mechanisms for heterogeneous graphs that differentiate importance across node and relation types dynamically.

Power Law Distribution Inference Methods

Develops statistical methods for identifying, fitting, and testing power-law distributions in network degree sequences and other properties.

Graph Neural Networks for Time Series

Applies graph neural networks to multivariate time series data where variables form networks, enabling spatio-temporal forecasting.

Influence Prediction in Social Networks

Predicts user influence, reachability, and impact potential in social networks using graph features and learned representations.

Network Slicing and Edge Computing

Optimizes network resource allocation and service provisioning in software-defined networks using graph partitioning and scheduling algorithms.

Entity Disambiguation in Knowledge Graphs

Resolves mentions of entities to canonical representations in knowledge graphs using graph structure and contextual information.

Meta-Learning for Graph Neural Networks

Develops meta-learning approaches enabling graph neural networks to quickly adapt to new tasks with few labeled examples.

Supply Chain Resilience Optimization

Analyzes and optimizes supply chain networks for robustness against disruptions, cascading failures, and external shocks.

Dynamic Community Structure and Stochastic Block Models

Advancing probabilistic generative models for time-varying community structure that capture latent block assignments and their temporal transitions in evolving networks.

Graph Foundation Models and Transfer Learning

Creating large-scale pre-trained graph models that transfer knowledge across diverse network domains and enable few-shot learning on downstream graph tasks.

Distributed Graph Processing Frameworks

Develops scalable distributed computing frameworks for graph processing, analysis, and machine learning on cluster and cloud infrastructure.

Fairness and Bias Mitigation in Graph Learning Systems

Investigating sources of algorithmic bias in graph neural networks and developing debiasing techniques to ensure equitable predictions across network subgroups and demographics.

Continuous-Time Graph Neural Networks

Develops neural ordinary differential equations and continuous-time models for graphs with irregular temporal sampling.

Interpretable Graph Patterns and Rule Mining

Extracting interpretable symbolic rules and patterns from complex graphs through frequent subgraph mining and logical reasoning over network structures.

Cross-Platform User Identity Linking

Links user accounts across multiple social platforms using network structure, features, and similarity metrics in multiplatform graphs.

Privacy-Preserving Differential Privacy in Graph Analysis

Designing differentially private algorithms for graph analytics that provide formal privacy guarantees while maintaining utility in network queries and machine learning tasks.

Cross-Domain Network Alignment and Matching

Developing algorithms to align and integrate multiple heterogeneous networks from different domains while resolving structural and semantic inconsistencies between graph representations.

Graph Regularization and Laplacian-Based Learning

Research on leveraging graph Laplacian matrices and spectral properties to design regularization techniques that enforce smoothness and structural consistency in semi-supervised and unsupervised learning tasks on networked data.