I'm a Computer Science graduate from Pennsylvania State University with a degree of Master in Science I have worked in related fields like deep learning in the NLP/NLG domain. For the past few years, I have been motivated by the following question: What are the powers of data-driven approaches in multidisciplinary domains of artificial intelligence? This question led to independent lines of work that include applying deep learning methods to solve research problems for healthcare analytics, text processing, and social network analysis. Meanwhile, I enjoy collaborating with researchers with different backgrounds on interdisciplinary research such as question answering, text summarization, health and social media analysis. Any kind of ideas and suggestions are welcome here.
My research papers are proceeding with NeurIPS MLPH 20, ICDAR 20, CODS COMAD 21, ICON 20, and Computational and Mathematical Organization Theory(Springer)
When I'm not working, I like playing soccer and listening to music. Currently, I have been looking for a full time role in the filed of Machine Learning and Natural Language Processing.
We aimed to mitigate the disparity of access to Telehealth among different racial groups. To design and experiment on different transformers based model like BERT, BART, and GPT2, we could generate a doctor response through question answering. The paper “Automated Medical Assistance Attention Based Consultation System” was accepted at NeurIPS MLPH workshop 2020.
We aimed at developing a classifier for fine-grained inference of COVID-19 tweets. Through building a benchmark system for misinformation mitigation and experimenting with classical machine learning approaches to the latest state-of-the-art deep learning models gave me a deeper insight into natural language processing. I presented the work “A Fine-Grained Analysis of Misinformation in COVID-19 Tweets” at the IdeaS conference which got accepted in a special issue of springer named “Computational and Mathematics Organization Theory”.
In this paper, we aim at Graph embedding learning for automatic grasping of low dimensional node representation on biomedical networks. The purpose is to use different neural Graph embedding methods for conducting analysis on 3 major biomedical link prediction tasks drug disease association (DDA) prediction, drug drug interaction (DDI) classification, and protein protein interaction (PPI) classification. We observe that graph embedding method achieve a promising result without the use of any biological features. The paper got accepted at CODS-COMAD 21 in Young Researchers Symposium Track.
In this paper, we have designed a character level pretrained language model for extracting support phrases from tweets based on the sentiment label. We also propose a character level ensemble model designed by properly blending Pre-trained Contextual Embeddings (PCE) models RoBERTa, BERT, and ALBERT along with Neural network models RNN, CNN and WaveNet at different stages of the model. The paper got accepted at 17th International Conference on Natural Language Processing [ICON 2020] .