California Housing Challenge 🏡

This educational challenge is to develop a machine learning model for predicting median block house prices in California using features such as the number of rooms and the age of the house.

This is a great opportunity to experiment with and learn about a number of core concepts in machine learning using pandas, seaborn and scikit-learn.

Getting started

To get started, check out our tutorial notebook:

https://github.com/DoxaAI/educational-challenges/blob/main/california-housing/getting-started.ipynb

The dataset

This challenge is based on the popular California housing dataset originally based on data from the 1990 US Census.

It contains the following data variables:

• median_income: the median income in block group in thousands of dollars
• house_age: the median house age in block group
• mean_rooms: the mean number of rooms per household
• mean_bedrooms: the mean number of bedrooms per household
• population: the block group population
• mean_household_size: the median household size of the block
• latitude: the latitude of the block group
• longitude: the longitude of the block group
• median_house_value: the median house value in thousands of dollars

If you have any questions about the challenge, feel free to reach out in the DOXA Community Discord server.