Bhumika

Bhumika

Senior Research Fellow
IIT Jodhpur
Email:bhumika.1@iitj.ac.in
VANET LAB 115, CSE, IITJ

I am a Phd Student in the Computer Science and Engineering Department at IIT Jodhpur. advised by Dr. Debasis Das. My research interests lie at the intersection of smart cities and artificial intelligence. In particular, the work so far has focused on the utilization of deep learning techniques in order to enhance the intelligence of the city. The work spans from designing spatio-temporal predictive models to responsible models.

Currently, I am designing defense mechanism to protect spatio-temporal learning systems against diverse attacks. My long-run goal is to leverage artificial intelligence for social good, particularly in shaping intelligent and socially responsible urban environments.

[ Curriculum Vitae ] [ Research Statement ]

Research Interests

  • Deep Learning
  • Spatio-temporal Learning
  • Federated Learning
  • Uban Computing
  • Responsible AI
  • Model Security
  • On-Device AI

Education

  • PhD in Computer Science & Engineering, 2020-Present

    Indian Institute of Technology Jodhpur, Rajasthan

  • MTech in Computer Science & Engineering, 2017-2019

    C-DAC, Noida affiliated to GGSIPU, Delhi

  • BTech in Computer Science & Engineering, 2013-2016

    ABVGIET, Pragatinagar affiliated to HPTU, H.P.

News

Received Registration Grant for attending CODS-COMAD 2025.
Our paper titled Co-Move: Covid-19 and Inter-region Human Mobility Analysis and Prediction has been accepted to IEEE TCSS.
Received a Registration Grant and served as a volunteer at PerCom 2024.
Received ACM India Research Facilitation Grant for Journal article.
Selected as Notable Participants in national level IITB-FOSSEE Mapathon.

Recent Publications

. Co-Move: Covid-19 and Inter-region Human Mobility Analysis and Prediction. IEEE Transactions on Computational Social Systems, 2024.

PDF Code

. Adaptive Context based Road Accident Risk Prediction using Spatio-temporal Deep Learning. IEEE Transactions on Artificial Intelligence, 2023.

PDF Code

. Federated learning based driver recommendation for next generation transportation system. Expert Systems with Applications, 2023.

PDF

Experience

University of Grenoble Alpes, France
Worked on distributed Spatio-temporal Learning with Prof. Phillipe Lalanda. We propose a Privacy-Preserving Edge-Based Urban Computing Architecture with Local and Global Spatio-Temporal Learning with autonomous model updates for spatio-temporal learning.
National Institute of Technology Hamirpur
I have taught courses on Game Theory and Wireless Sensor Networks for Dual Degree students. Additionally, I have conducted undergraduate Computer Graphics lab sessions, guiding students through hands-on implementation of core graphics concepts.

Projects

*
Traffic Hotspot Analysis
Jodhpur, the second-largest city in Rajasthan, faces increasing urban traffic challenges. In this project, we analyzed traffic hotspots by extracting road network and traffic data specific to Jodhpur city. The data was visualized and mapped using QGIS software, allowing for a detailed spatial analysis. The results highlight key congestion zones in central areas such as Nai Sadak Road, Sardarpura B Road, Neem ki Gali, and Residency Road. This project was carried out as part of the IIT Bombay–FOSSEE Mapathon 2023, contributing to open-source geospatial mapping and urban mobility research.
Traffic Hotspot Analysis of Jodhpur City

Activities \ Grants

Received registration grant for attending CODS-COMAD 2025.
Received registration grant for attending PerCom 2024.
ACM India research facilitation grant for extra page charges for Journal article.