Currently, I am working as a Senior Data Scientist in a financial services company. My work is focused on developing financial prediction models for existing and prospective clients. Prior to this, I was working as a post-doctoral fellow at the University of Houston after having completed my PhD in Computational Science at the University of Southern Mississippi. Data Science, Machine Learning, Data mining, and related techniques play a pivotal role in my research on medical image processing for diabetic retinopathy, information diffusion in social media, Goal-oriented Chabot, and Theory of Mind.

Research Projects

SpaceML: Technical Team Member

1. Working on the NASA Meteor Shower web portal project, which helps the user to browse through aggregated meteor, star, and constellation data from different stations in the past few years.
Please visit: https://meteorshowers.seti.org
2. Contributed to CAMS-Net Meteor Detector ablation analysis, implementing deep learning models and attention mechanism.

PATENTS

1. “Watchdog event for blocking unknown event while accessing Big Data” (South Korea Patent KR1020140089282).

2. “Display of important area on zooming for Smart TV”(South Korea Patent KR1020150105778).

3. “Zooming beyond the physical screen with AR compatible devices” (South Korea Patent KR1020160125027).

TECHNICAL TALKS:

[1] Topic: Important Mathematical Techniques for Machine Learning, at CANTOR SECT, Department of Mathematical Sciences, The University of Memphis, Memphis, TN 3815. [PPT]

[2] Topic: Important Mathematical Techniques for Machine Learning, at Department of Computer Science and Engineering & Information Technology with Institution's innovation council of Bengal Institute of Technology, Kolkata, Tech Town, Basanti Highway, Bantala (Bus-stop: 1 No. Colony ), West Bengal, India, Kolkata-700150. [PPT]

PUBLICATIONS:

[12] Claim Verification using a Multi-GAN based Model, RANLP 2021: Recent Advances in Natural Language Processing (Online event). https://arxiv.org/abs/2103.08001

[11] On the Feasibility of Using GANs for Claim Verification-Experiments and Analysis, Workshop on Reducing Online Misinformation through Credible Information Retrieval, April 2021, Tuscany, Italy (Online event).

[10] Early Detection of Diabetic Retinopathy from Big Data in Hadoop Framework, Displays - Journal – Elsevier.

[9] Information Diffusion on Twitter: Pattern Recognition and Prediction of Volume, Sentiment, and Influence, Fourth IEEE/ACM International Conference on Big Data Computing, Applications and Technologies, ACM, p157-167, December 2017, New York, USA.

[8] Goal-Oriented Conversational System Using Transfer Learning and Attention Mechanism, The 10th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference, Columbia University, October 2019, New York City, USA.

[7] Dialogue Generation Using Self-Attention Generative Adversarial Network, First IEEE International Conference on Conversational Data & Knowledge Engineering (CDKE 2019), October 2019, San Diego, USA.

[6] Influence Modeling, Volume Prediction and Sentiment Analysis of Short Texts on Twitter, Proceedings of the 31st International Conference on Computer Applications in Industry and Engineering (CAINE’18), 2018, New Orleans, USA.

[5] User Level Weighted Multi-Feed Topic Embeddings for studying Network Interactions in Twitter, International Conference on Big Data, June 2019, San Diego, USA.

[4] Collective Decision Making by Swarm Robotics Using Blockchain, International Congress on Blockchain and Applications, in June 2019, Avila, Spain.

[3] Building a learning machine classifier with inadequate data for crime prediction, Journal of Advances in Information Technology Vol 8.2 (2017).

[2] How the Tables have Turned: Studying the New Wave of Social Bots on Twitter Using Complex Network Analysis Techniques, IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), August 2019, Vancouver, Canada.

[1] Optimal Feature Selection from VMware ESXI 5.1 Feature set, in Academy & Industry Research Collaboration Center, September 2014.

IMPORTANT PREVIOUS PROJECTS

1. Developed a system to Identify signs of diabetic retinopathy in eye images using Hadoop, MapReduce, and Mahout framework.

2. Developed a voice data analysis system for the Customer care centers using Hadoop and MapReduce framework.

3. Developed a Log analysis system using Hadoop and MapReduce framework for an analysis of user activities.

4. Road direction in camera preview of Android smartphone Proposed the idea. Designed and developed the algorithm as part of a five-member. Integrating Google Map APIs and LG Camera application.

5. Camera flash reduction (Collaboration with Indian Institute of Science, Bangalore) Determining the flashlight effects in the image taken by a mobile phone’s camera and reducing white patches. Worked on designing the algorithm and implementation. Mainly using Poisson blending

6. Structured data retrieval Proposed the idea. Designed and developed the algorithm as part of a five-member team. Integrating Google Map APIs and LG Camera application.

M.Tech MAJOR PROJECT

Clustering of Virtual Machines Depending on Workloads Using Unsupervised Learning Algorithms-Analysis of different loads for system resources like memory, processor, I/O devices etc. Used various feature selection algorithms for the analysis of data. Used Kmeans and EM algorithms for clustering the virtual machines. Used Davies-Bouldin index and Dunn index for cluster analysis.[pdf ]

INTERNSHIP:

In Indian Statistical Institute, Kolkata(June 2012 - August 2012)

Implemented some of the basic tasks images, Enhancement, and image segmentation.