Advancing unsupervised graph clustering, incremental learning algorithms, and adaptive anomaly detection. Developing scalable machine learning solutions for complex networks and dynamic data systems.
Assistant Professor
Department Of Computer Science and Engineering
National Institute Of Technology, Rourkela
Developing self-supervised GNN architectures for unsupervised graph clustering and community detection in complex networks using deep learning and contrastive learning approaches.
Designing adaptive incremental clustering algorithms for dynamic datasets including BISDBx, iMass, and BiMass using density-based approaches for real-time data stream analysis.
Creating intelligent outlier detection frameworks using kernel density estimation and grid-based methods for identifying anomalies in evolving data streams and large-scale systems.
Advancing unsupervised graph clustering through stable GNN architectures. Developed G-DMoNLite for efficient community detection with improved training stability and reproducibility.
Enhancing Graph Clustering framework through curriculum learning that gradually transitions from structural to semantic optimization, combined with entropy-based confidence weighting.
Developing modularity-based loss functions that jointly optimize graph structure and node attributes. Created attributed modularity loss to improve clustering quality by integrating topological and feature information in neural graph clustering frameworks.
13th International Conference on Data Science (ACM IKDD CODS 2025)
Click here13th International Conference on Data Science (ACM IKDD CODS 2025)
Click here13th International Conference on Data Science (ACM IKDD CODS 2025)
Click herebhattacharjeep@nitrkl.ac.in
panthadeep.edu@gmail.com
Room Number: CS-320
Department of Computer Science and Engineering
National Institute of Technology Rourkela
Odisha, India - 769008