Projects, articles, research and activities by students at Notre Dame College, Dhaka.
Project
Personally handled most project deliverables, including report writing, visuals, presentation materials, and assembly of a working AWG prototype. Contributed to planning, cost analysis, and feasibility assessment; worked with Peltier modules, cooling fans
Project
Surveyed 30 respondents through Google Forms; applied the Multi-Attribute Attitude Model to compare Maggi, Mr. Noodles, Doodles, and Cocola on taste, price/value, healthiness, and availability; contributed to calculations and report writing. Mr. Noodles r
Project
In a five-member team, surveyed 31 patients using the SERVQUAL model (Tangibles, Reliability, Responsiveness, Assurance, Empathy) on a 7-point Likert scale; organized data, contributed to SPSS analysis, and designed presentation slides. Refined the questi
Project
Built an Arduino robot with beacon homing, obstacle avoidance and sensor-based sorting of waste.
Project
Built a Flask website as a group project for cataloguing, borrowing and managing books.
Project
Built a Django platform where passengers find and book bus rides and drivers add and manage trips. Added two-factor authentication and profile management for safe and simple use.
Project
Built a full-stack MERN web app for pet adoption as a team project, using MongoDB Atlas and Git.
Project
Built an end-to-end workflow covering EDA, cleaning, encoding, scaling and supervised modeling. Handled class imbalance and compared models on accuracy, precision, recall and ROC-AUC, with a KMeans baseline.
Project
Built a pipeline to classify breast cancer samples from high-dimensional RNA-seq data (TCGA-BRCA). Used log transform and mutual-information feature selection, then compared Logistic Regression, XGBoost and a Neural Network under class imbalance.
Project
Evaluated Platt scaling and isotonic regression for tree-based models on binary and multiclass healthcare tasks. Used the CDC Diabetes Health Indicators 2023 data with ROC-AUC, Brier score, ECE and reliability diagrams.
Project
Deep generative framework to integrate RNA, ATAC, proteomics and metabolomics data where each dataset has only some modalities. Uses modality-specific encoders and decoders with a shared latent space, plus missing-not-at-random (MNAR) modeling. Adds calib
Project
It is a short film on gender discrimination