HI, I AM

Chaity Rani Ghosh

Researcher in Human-Computer Interaction and Machine Learning.

Computer Science & Engineering graduate from BRAC University and Student Researcher at ELITE Research Lab. Specialized in constructing trustworthy deep learning models, multimodal diagnostic pipelines, and transparent medical decision support architectures.

Chaity Rani Ghosh

Background & Academic Mission

I am a Computer Science and Engineering graduate from BRAC University, now conducting research at the master's level while serving as a Student Researcher at ELITE Research Lab. My academic focus lies at the intersection of explainable AI (XAI) and human-AI interaction (HAI)—from synthesizing multimodal clinical diagnostics (radiographs, EHR notes, knowledge graphs) to studying how humans develop cognitive and emotional trust in conversational AI.

The core motivation across all my investigations is simple: how can we design machine learning architectures that are not only statistically accurate, but explainable, accountable, and reliably aligned with human experts? I am currently preparing applications for PhD programs in Computer Science to continue this research agenda.

3.77 CGPA / 4.00 (High Distinction)
5 Active Research Projects
7 Semesters on Dean's & VC's List
15% Exam Improvement as TA (CSE330)

Tools & Methodologies

Technical proficiencies used across machine learning pipelines, systems, and scientific publications.

Languages

Python Java C / C++ SQL PHP JavaScript HTML5 / CSS3

ML, Deep Learning & Vision

PyTorch TensorFlow Scikit-learn Hugging Face DenseNet121 ResNet50 BERT VLMs XGBoost

Explainability & Vision (XAI)

Grad-CAM LIME Attention Heatmaps Knowledge Graphs Feature Attribution

Scientific Computing & Data

NumPy Pandas Matplotlib Seaborn Hypothesis Testing Cross-Validation

Systems, Networking & Tools

Docker Git & GitHub Linux / Bash Ryu SDN Controller Mininet MySQL Laravel

Academic Standards & HCI

LaTeX / Overleaf IEEE / ACM Standards Mixed-Method HCI Survey Protocol Design Literature Reviews

Research & Work Experience

Sep 2026 — Present Research

Student Researcher

ELITE Research Lab · Queens, NY, USA (Remote)
  • Investigating Human-AI Interaction (HAI) utilizing Natural Language Processing (NLP) and Large Language Models.
  • Designing evaluation protocols to assess usability, transparency, and calibrated user trust in AI interfaces.
Jun 2025 — Dec 2025 Engineering

Technology Intern

Technology Unit, BRAC International · Dhaka, Bangladesh
  • Conducted functional testing, debugging, and systematic documentation of enterprise software tools.
  • Assisted in maintenance and deployment of internal technology systems and services.
Oct 2023 — Jan 2025 Academic

Undergraduate Teaching Assistant

Dept. of Computer Science & Engineering & MNS, BRAC University
  • Delivered problem-solving tutorial sessions for Numerical Methods (CSE330) and Complex Variables & Fourier Analysis (MAT215).
  • Mentored students through complex theoretical derivations, helping elevate cohort exam scores by 15%.
  • Evaluated and provided detailed technical feedback on 300+ assignments per term.

Academic Qualifications

Completed 2021 — 2025

Bachelor of Science in Computer Science and Engineering

BRAC University · Dhaka, Bangladesh
CGPA: 3.77 / 4.00 High Distinction Graduate

Received Dean's List and Vice Chancellor's List honors across 7 academic semesters. Completed undergraduate research thesis on Software Defined Network load balancing using machine learning.

In Progress Present

Master of Science in Computer Science and Engineering

Graduate Studies · Dhaka, Bangladesh

Advanced coursework and thesis research focused on multimodal learning, explainable medical diagnostics, and trust calibration in intelligent systems.

Selected Technical Projects

Open-source implementations, benchmarks, and machine learning pipelines.

2025

Coronary Heart Disease Risk Predictor

Clinical predictive modeling tool utilizing a hybrid ensemble to predict 10-year risk of coronary disease with automated hyperparameter tuning and cross-validation.

Python Scikit-learn NumPy Pandas
2025

Handwritten Bangla Name Recognition & XAI

Computer vision transfer learning pipeline on custom self-curated dataset, featuring ResNet50 (83.89% test accuracy), Grad-CAM heatmaps, and local LIME explanations.

TensorFlow ResNet50 Grad-CAM LIME
2024

Healthy Habits — Diet Planning Platform

Full-stack MVC web platform enabling users to track, schedule, and plan nutrition with relational schema validation and caloric calculators.

Laravel PHP MySQL JavaScript
2024

Diagnostic Diabetes Predictor

Supervised classification pipeline comparing logistic regression, SVM, and random forests on clinical metabolic markers.

Python Scikit-learn Matplotlib

Publications & Manuscripts

Peer-reviewed conference proceedings, journal papers under review, and preprints.

IEEE AIMLA 2025 2025

Hybrid Q-Learning with VLMs Reasoning Features

Ashraf, M. S., Akuthota, V., Prapty, F. T., Sultana, S., Riad, J. A., Ghosh, C. R., Hasan, N., & Anwar, A. S.

2025 3rd International Conference on Artificial Intelligence and Machine Learning Applications (AIMLA), IEEE.

Under Review 2025

Time-Sliced Round Robin (TSRR) Scheduling for Flow-Aware Load Balancing in Software Defined Networks

Ghosh, C. R. · Supervisor: Dr. Jannatun Noor Mukta

Under Review at Alexandria Engineering Journal (Elsevier).

Introduces an adaptive traffic scheduler for Software Defined Networks combining machine learning flow classification with dynamic time slicing, achieving 91% traffic classification accuracy, 85% jitter reduction, and 89% lower packet loss over static baselines.

Preprint 2026

Explainable Multimodal Chest X-ray Diagnosis Using Clinical Text and Medical Knowledge Graphs with VLM-based Reasoning

Ghosh, C. R. · Supervisor: Dr. Md. Ashraful Alam

Manuscript / Working Paper

Multimodal fusion architecture coupling DenseNet121 visual features, clinical BERT text embeddings, and structured medical graph priors, interpreted using Grad-CAM heatmaps.

DenseNet121BERTKnowledge GraphsGrad-CAM
Preprint 2026

Machine Learning-Based Coronary Heart Disease Prediction: A Comprehensive Ensemble Approach

Ghosh, C. R. · Independent Research

Manuscript / Working Paper

Ensemble framework integrating Decision Trees, Random Forests, and XGBoost with automated feature selection for coronary risk prediction.

XGBoostEnsemble LearningFeature Selection
Preprint 2025

Being Heard Without Being Connected: Emotional Dependence on AI Chatbots and Its Perceived Implications for Mental Well-being among University Students

Ghosh, C. R. · Supervisor: Dr. Jannatun Noor Mukta

Manuscript / Working Paper

Mixed-methods empirical study assessing anthropomorphic attachment, conversational reliance, and ethical implications of conversational LLMs.

HCIMixed MethodsAI Ethics
Preprint 2025

Robust Handwritten Bangla Human Name Recognition Using Deep Learning, Ensemble Models and Explainable AI

Ghosh, C. R. · Supervisor: Mr. Annajiat Alim Rasel

Manuscript / Working Paper

Transfer learning framework evaluated on a curated handwritten benchmark, achieving 83.89% test accuracy with ResNet50 and audited via LIME and Grad-CAM.

ResNet50LIMEGrad-CAM
B.Sc. Thesis 2025

An SDN-based Approach Using RYU Controller for Load Balancing and Performance Evaluation in Hybrid Networks with Machine Learning Algorithms

Ghosh, C. R., Ahsan, N., Islam, M., Noor, R., & Mashrafi, A.

Bachelor's Thesis, Dept. of Computer Science & Engineering, BRAC University, 2025.

Honors & Certifications

Academic Honor

Dean's List & Vice Chancellor's List

Recognized for outstanding academic merit in 7 semesters at BRAC University.

Graduation Honor

High Distinction Graduate

Graduated with CGPA 3.77 / 4.00, B.Sc. in Computer Science & Engineering.

Certification

Understanding Machine Learning

DataCamp (2025) — Machine learning paradigms, validation, and optimization.

Certification

Introduction to SQL

DataCamp (2025) — Relational querying, complex joins, and aggregations.

Editorial Review

Manuscript Under Review

TSRR scheduling architecture at Alexandria Engineering Journal (Elsevier).

Let's Discuss Research or PhD Opportunities

I am actively exploring PhD openings and research collaborations in Multimodal Machine Learning, Medical AI, and Explainability. Feel free to reach out directly:

Citation copied to clipboard