This repo explores KMeans and Agglomerative Clustering effectiveness in simplifying large datasets for ML. Goals include dataset download, finding optimal clusters via Elbow and Silhouette methods, comparing clustering techniques, validating optimal clusters, tuning hyperparameters. Detailed explanations and analysis are provided.

Full-stack Todo application developed as part of a thesis on frontend frameworks. Built with React, Node.js, and PostgreSQL, it features user authentication, CRUD operations, and a responsive UI. Demonstrates modern web development practices, API integration, and serves as a practical case study for comparing frontend technologies.

This repository contains a Machine Learning project, implemented in python, named 'Crop Recommendation System using Extreme Learning Machine'.

Investigate the feasibility of machine learning procedures in predicting the fake news and misleading information affecting the image of social media company.Tasks are to identify machine learning techniques appropriate for a particular practical problem,undertake a comparative evaluation of several machine learning procedures when applied to the…

Comparative-Models for MNIST Dataset

This project explores the rapid growth of Buy Now, Pay Later (BNPL) services and their impact on consumer spending habits and credit risk in the fintech sector. It aims to assess how BNPL influences financial behavior, debt management, and the evolving landscape of digital payments.

SWEETGUARD 🛡🔍 – A data-driven diabetes risk assessment tool that leverages machine learning and public health datasets to predict individualized diabetes risk scores. Using Python 🐍, Power BI 📊, and statistical analysis, this project identifies key lifestyle factors and empowers individuals with personalized health insights.

Research study about business strategies: profit driven and growth driven

XAI LIME-based image classifier or SHAP unified framework is utilised to interpret the output of different types of models, such as neural networks, decision trees, and linear models, is to FGSM adversarial assaults and evaluate the effect on the predictability of the model.