PB

Panthadeep Bhattacharjee

National Institute of Technology Rourkela, Indian Institute of Information Technology Guwahati, Indian Institute of Technology Guwahati, National Institute of Technology Durgapur, Assam University

ORCID iD 0000-0002-2849-2161
Follow

About

PhD. (CSE/IIT Guwahati) | M.Tech (CSE/NIT Durgapur) | B.Tech (CSE/Assam (Central) University, Silchar) | Faculty (CSE/NIT Rourkela) | Faculty (CSE/IIIT Guwahati) | Got selected for a Post-Doc in 2022 at the University of Porto, Portugal, for working on a project titled "Generating Artificial Data for ML algorithms"

I focus on developing incremental data mining algorithms with an emphasis on Clustering and Anomaly detection. Recently started working on Deep Graph Clustering; Areas of work: Data Mining;
Machine Learning; GNNs. Key publication venues so far: ACM SIGIR, ACM SIGAPP/SAC, ECIR, ACM IKDD-CODS, ICDCIT, ACIIDS, AIAI, PAAA (journal), FCS (Journal).

I have primarily worked on designing incremental algorithms for handling dynamic data for mining clusters and outliers. In the course of my research until now, I have developed four incremental
mining algorithms, out of which three are related to clustering, and one is an outlier detection algorithm. Among the four algorithms, two are exact incremental, and the remaining are approximate incremental
algorithms. The updates to these methods are made in batches as well as point-based. While designing incremental algorithms, it is important to know about the nature of the state-of-the-art before we attempt
to design its incremental version. To the best of our observation, there is no standard framework to produce the dynamic extension of the naive methods. The heavier components of the benchmark are recalibrated to reduce the time complexity, while the less intensive components are kept intact. I have also conducted one of the most comprehensive studies about density-based clustering algorithms that were proposed since 1996, ranging over a period of 23 years.

As a part of my current research, I am studying various techniques to find communities within complex networks. I am also working on predicting priority-based layer-wise connections in social networks using density-based clustering. Our group is currently working on GNN-based clustering and Contrastive Learning. We have been able to make certain inroads regarding that, on top of some of the recent works. I am also collaborating with my colleagues from other institutes, where I can contribute through my expertise in their domain of work

Employment

  • National Institute of Technology Rourkela Teaching, Research
    2023 - Present
  • Indian Institute of Information Technology Guwahati Teaching, Research
    2022 - 2023
  • Kalinga Institute of Industrial Technology Teaching, Research
    2020 - 2022
  • Kalinga Institute of Industrial Technology Teaching, Research
    2012 - 2013

Education

  • Indian Institute of Technology Guwahati PhD.
    2014 - 2020
  • National Institute of Technology Durgapur M.Tech
    2010 - 2012
  • Assam University B.Tech
    2006 - 2010
  • Cotton University Higher Secondary (10+2)
    2003 - 2005
  • Happy Child High School Matriculation (10th)
    1995 - 2003

Projects & Funding

Projects & funding information is unavailable.

Publications (27)