Selected work

A mix of machine learning, web applications, and practical tools—each one built to solve a specific problem.

Machine learning 01

Handwritten digit recognition

Trained a convolutional neural network on the MNIST dataset to identify handwritten digits, then built a web app to interact with the model.

The model reached 98.63% accuracy on the MNIST dataset.

  • Python
  • Pandas
  • TensorFlow
  • Keras
  • NumPy
  • Matplotlib
View source on GitHub

Web application 02

Recipe recommender

A web app that helps people find dishes based on the ingredients they have. Neo4j models the relationships between ingredients and recipes.

  • Neo4j
  • Node.js
  • Express
  • W3.CSS
View source on GitHub

Computer vision 03

Pulmonary disease detection

Explored feature extraction and broad disease classification using lung X-rays and CT scans. The project focused on identifying and grouping potential pulmonary diseases from medical images.

  • TensorFlow
  • scikit-learn
  • OpenCV

Web application 04

Child day care system

A web-based enrollment system designed to help parents register children for daycare without waiting in a queue. An admin dashboard summarizes babysitters, services, and enrollment status.

  • PHP
  • HTML
  • CSS
  • MySQL