Machine Learning

Machine Learning is a powerful tool for solving complex problems in a variety of domains. Here are some of my previous machine learning projects, which include recipe recommendation systems and classification of vehicular collisions.

Recipe Recommendation System

This project involved the creation of a recipe recommendation system using a dataset of 1000 recipes. The dataset contained information on recipe category, nutritional information, and serving size. The final model was able to recommend popular recipes 81.4% of the time, 34.4% improved over conventional methods.

Classification of Vehicular Collisions

This project involved the classification of vehicular collisions in the Chicago Municipal Area; The dataset, sourced from the Chicago Open Data Portal, contained over 1.5 million records and 29 features. The final model was able to predict the likelihood of a service callout with a recall rate of 88.7%.

Deep Learning

Deep Learning is a subset of machine learning that uses neural networks to model and understand complex patterns in data. Here are some of my previous deep learning projects, digit recognition and malaria detection.

SVHN Digit Recognition

This project involved the classification of house numbers in the Street View House Numbers (SVHN) dataset. The dataset contained over 600,000 images of house numbers from Google Street View. The final model was able to predict the correct house number with 91% accuracy and F1 score.

Malaria Detection

This project involved the classification of malaria-infected cells in the Malaria Cell Images dataset. The dataset contained over 27,000 images of malaria-infected and uninfected cells. The final model was able to predict the correct cell type 99% of the time, 4% improved over industry standards.