ABOUT ME

I am a PhD Candidate at Computer Science department of Lehigh University and working at DAS lab under Professor Eric P.S. Baumer supervision. I also hold industry research experience of 10-month internship working at Dana-Farber Cancer Institute, and 3-month as PhD NLP researcher internship at Infinitus Systems.

I am interested in improving AI techniques to align more with human understanding, enhancing current machine learning assessment approaches, and exploring human-computer interactions. I am also interested in using NLP methods to improve human-computer interaction; I adopt NLP techniques to explore a corpus or to generate outputs to help human users save time and access large number of documents faster.

Bias as a Distinct Factor in Human Ratings of Machine Labeling

Human assessments of machine labeling can reveal bias as a distinct measure separate from other perceptions of quality.

One Rating to Rule Them All?

human assessments of labeling quality have a two-factor latent structure. Subsequent analysis demonstrates that this multi-item, two-factor assessment can reveal nuances that would be missed using either a single-item human assessment of perceived label quality or established performance metrics.

Text2table

Text2Table is a two part package for fast and reliable clinical named entities and relation extraction. First part, LinearNER, includes a very quick and easy to deploy approach to extract named entites. The second part, AINER, detect named entities and relation extraction between the extracted named entites using deep learning transformers.

News
June 5, 2023: I will join Infinitus Systems as Phd NLP Researcher starting June 19 for 12-week internship
May 22, 2023: Our paper, An Interdisciplinary Approach to Understanding Cultures of Ethics in STEM, got accepted at BSTS.
July 20, 2022: Our paper, One rating to rule them all?, got accepted.

What I am doing in a nutshell

Human-Centered AI

I like to improve Human-AI interaction experience by providing further information to human users

AI Assessment

First thing to to provide a better environment for human users when working with machine learning tools is to evaluate machine the way humans do. Improving AI assessment should not be restricted to good or bad, but to multiple criteria human users may consider.

Topic Modeling

I am using topic modeling to statistically analyze a corpus, explore documents, and assist qualitative coding to interpret main themes better. I am also interested in improving topic modeling evaluation and performance.

Deep Learning NLP

I enjoy Working with transformers and BiLSTM to improve NLP feature extraction and classification.