This project involved analyzing data from a food delivery app called Foodhub. The data consisted of information about nearly 2,000 food orders, including the cuisine type of each order, the delivery time, rating, restaurant name, and more. Using NumPy, Pandas, and Seaborn libraries, my goal was to draw insights from the data, visualize trends, and ultimately offer recommendations to improve the app and business.
The goal for this project was to build a classification machine learning model that predicts whether a customer “lead” will become a paying customer for the online learning platform ExtraaLearn. Using data from over 4,000 profiles, I built, trained, and tuned both a decision tree model and random forest model with Scikit Learn and GridSearch.
For this project I built several machine learning models for Amazon that recommend products to customers. The first model is a “rank based” recommendation system which processes the rating data and recommends the top 5 most popular products based on their average rating and number of ratings. The next three models I built are collaborative filtering recommendation systems - these are more personal to the user. These models include a user similarity based system, an item similarity based system, and lastly, a model built and trained by using a singular value decomposition of the user/item matrix.
