Projects Portfolio
This project applies machine learning techniques to classify ovarian cancer stages, aiding in early detection and personalized treatment strategies. Using patient data and biomarkers, we train predictive models to identify cancer subtypes accurately.
This project employs deep learning techniques for bird species classification, leveraging convolutional neural networks (CNNs) to analyze and identify various bird species from images. The model is trained on a diverse dataset, enabling accurate species recognition and contributing to ecological research and conservation efforts.
Streamlit-powered YOLO license plate detection app, offering image/video processing with interactive user interface, allowing uploads for detection. Training data gotten from Roboflow. The project processes both images and videos, offering versatile use cases.
A concise machine learning project leveraging Logistic Regression, SVM, and Decision Trees to classify chemical compounds. Incorporates cross-validation, hyperparameter tuning, and feature
selection for performance optimization on the high-dimensional, class-imbalanced Dorothea dataset.
This repository contains a data-driven project analyzing various factors influencing unemployment rates across African countries. The project delves into regional disparities, gender differences in employment, the impact of education funding, private sector growth, and infrastructure development on job markets.
Inspired by a Premium Times article on data privacy breaches by digital lenders in Nigeria, this project uses machine learning for sentiment analysis of app reviews from six platforms (Quickcredit, Carbon, Newcredit, Fairmoney, Branch, Palmcredit) on Apple and Google Play stores. It leverages natural language processing to gauge customer sentiments towards these digital lending services.
This is a data analysis project focused on improving health outcomes in Africa. Through data exploration and modeling, the project aims to provide insights and recommendations to address pressing health issues. The repository includes code, datasets, and findings and welcomes contributions from the open-source community.
This project focuses on detecting fraudulent credit card transactions using Kaggle’s simulated dataset. It employs a random forest model, with fine-tuned hyperparameters, to differentiate legitimate from fraudulent transactions effectively