Mini-projects undertaken during the "INFO 251: Applied Machine Learning" course at UC Berkeley.
Each of folders in this repository contains a separate mini-project. For the most part, the included Jupyter notebooks are well-documented and should be understandable by most audiences.
-
PS1
- A very basic exploration of the Boston Housing Prices dataset.
-
PS2
- Hypothesis tests to evaluate the impact of the Progresa program - a government social assistance program in Mexico.
-
PS3
- Using k-nearest neighbours to predict the median value of owner-occupied homes in the Boston Housing Prices dataset.
-
PS4
- Similar to PS3, but using gradient descent linear regression with regularization.
-
PS5
- Using a naive Bayes classifier to predict whether a movie is 'fresh' or 'rotten' on Rotten Tomatoes.
-
PS6
- Using decision trees and random forests to predict survival on the Titanic.
-
PS7
- Using k-means clustering and principal components analysis to recognize faces in the Faces in the Wild dataset.