Anuska Maity

MS (Research) Student | Natural Language Processing & Cognitive Science Researcher
Kolkata, IN.

About

Highly motivated MS (Research) student specializing in Natural Language Processing and its intersection with cognitive science, with a strong foundation in data analysis and deep learning. Eager to leverage expertise in NLP techniques and data-driven insights to solve complex real-world problems and drive innovation in research and engineering roles.

Experience

Applied Research Works Inc. – Cozeva
|

Data Analyst

Kolkata, West Bengal, India

Summary

Analyzed patient data to enhance the end-user experience of the 'Cozeva' platform and facilitate efficient care gap closing for Quality & Risk Metrics and Data Bridge Projects.

Highlights

Optimized the 'Cozeva' platform's end-user experience by computing patient data for Quality & Risk Metrics, directly supporting data-driven healthcare decisions.

Contributed to Data Bridge Projects, streamlining the process of care gap closing and improving operational efficiency.

Utilized data analysis techniques to translate complex patient information into actionable insights for platform enhancements.

Internship

Centre for Digital Technologies in Healthcare (CDiTH)
|

Research Intern

Hyderabad, Telangana, India

Summary

Conducted in-depth analysis of mammogram data reports to develop an NLP-based model for automated BI-RADS score prediction, contributing to early cancer detection research.

Highlights

Analyzed extensive mammogram data reports from Grace Cancer Foundation and Yashoda Hospitals to identify key linguistic patterns for medical diagnostics.

Developed an NLP-based model designed for automated BI-RADS score prediction, aiming to enhance diagnostic accuracy and efficiency in breast cancer screening.

Applied advanced natural language processing techniques to extract and interpret critical information from unstructured medical text data.

Education

IIIT Hyderabad
Hyderabad, Telangana, India

MS (By Research)

Computer Science and Engineering

Grade: 8.33/10.0

Courses

Introduction to Cognitive Science

Behavioral Research: Experimental Design

Learning and Memory

Statistical Methods in AI

Introduction to NLP

Heritage Institute of Technology
Kolkata, West Bengal, India

B.Tech

Electronics and Communication Engineering

Grade: 8.78/10.0

Hem Sheela Model School
Durgapur, West Bengal, India

Higher Secondary (AISSCE)

Science

Grade: 93.6%

Hem Sheela Model School
Durgapur, West Bengal, India

High School (AISSE)

Science

Grade: 10.0/10.0

Awards

2nd Place, Inter-College Solo Eastern Vocals

Awarded By

Medical College Kolkata

Secured 2nd place representing Heritage Institute of Technology in a prestigious inter-college vocal competition.

Publications

Comparing Cue and Target Roles Reveals Limits of Intrinsic Word Memorability in Cued Recall

Summary

Lead author on a paper on investigating how word memorability varies by functional role within specific memory tasks, utilizing large-scale dataset analysis and reliability measures. Under Review : Journal of Memory and Language (Elsevier). Preprint Link (PsyArxiv) added.

Languages

English
Hindi
Bengali (Native)

Skills

Programming

Python, SQL, C++, C.

Frameworks & Libraries

PyTorch, Hugging Face Transformers, Scikit-learn, Pandas, NumPy.

Research & Analysis

Statistical Analysis, Literature Review, Academic Writing, Natural Language Processing (NLP), Deep Learning, Data Analysis, Machine Learning, Cognitive Science, Experimental Design.

Interests

Creative Pursuits

Singing, Writing, Band Performing (Vocalist).

Culinary Arts

Cooking.

Projects

Intricacies of Word Memorability | MS Thesis Project (ongoing)

Summary

Ongoing MS Thesis Project focused on analyzing how word memorability varies by functional role within specific memory tasks, contributing to the field of cognitive science.

Enhancing Intent Classification with Regularized Transformers | MS Course Project

Summary

Explored and compared the performance of advanced HuggingFace transformers with regularization techniques for banking intent classification.

Badminton Shot Classification using Spatiotemporal Deep Learning | MS Course Project

Summary

Developed and compared spatiotemporal deep learning models for accurate badminton shot classification using video-extracted image data.