Currently I am a PhD scholar in Computer Science and Engineering Department of Indian Institute of Technology, Kharagpur. I joined Ernst & Young (EY) as a data scientist and worked on various NLP and deep learning projects from 2019-2021. My current reasearch interest includes NLP, AI, Computer Vision and deep learning. I have completed MS (by research) under the guidance of Bivas Mitra and Niloy Ganguly in 2018 from the same institute. My broad area of research in MS was Anomaly Detection, Root Cause Analysis. I am also a member of Complex Networks Research Group, IIT, Kharagpur.
Education
Data Scientist in EY, 2018-2021
MS( by research)Indian Institute of Technology, Kharagpur, 2015-2018
B.Tech from kalyani Govt. Engineering College, 2010-2014
H.S from Baita Mahendra Nath High School, 2010
Teaching Assistanceship
| Id | Subject |
|---|---|
| CS19001 | Programming and Data structure lab |
| CS60050 | Machine Learning |
Research Projects
My research interests are in the domain of NLP, AI, CV, and deep learning, log analytics, anomaly detection. During my MS tenure, I have worked on Anomaly detection from Event Log Empiricism.
In this project we aim to detect anomaly from a large scale systems using EMS (Event Message System) log . Also developed an automated early warning system to protect the system from failure.
I have also developed an unsupervised troubleshooting methodology for identifying problematic modules responsible for system failure by constructing a sequence of time evolving dynamic graphs from the collected system log of customer filed cases. The framework reduces the complexity of the diagnostic process.
Publications
Subhendu Khatuya, Jayanta Basak, Niloy Ganguly, Madhumita Bharde, Bivas Mitra, ADELE: Anomaly Detection from Event Log Empiricism , In Proceedings of IEEE International Conference on Computer Communications (INFOCOM), Honolulu, HI, USA, 15-19 April 2018 (Accepted) (Acceptance rate: 19.2%) PDF
Subhendu Khatuya, Ajay Bakhshi, Jayanta Basak, Niloy Ganguly, Bivas Mitra, GBTM: Graph Based Troubleshooting Method for Handling Customer Cases Using Storage System Log, The 22nd Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD), Melbourne, Australia 2018 (Accepted) (Acceptance rate:17.79%)
Contact Me
Complex Network Research Group (CNeRG) LabRoom No: 205 Department of Computer Science and Engineering Indian Institute of Technology, Kharagpur mail: subha.cse143[AT] gmail [DOT] com