Predict the poverty of households in Costa Rica using automated feature engineering

Social programs have a difficult time determining that the right people are given enough aid. Using a dataset of Costa Rican household characteristics, we'd like to be able to predict the poverty of households.

We will show how Featuretools can be used to predict the poverty of household in Costa Rica using a dataset from Kaggle.

The Tutorial notebook from this repository exists on Kaggle. If you would prefer to work in that environment, you can fork the existing kernel to use as a starting point.

Highlights

• Automatically generate 2000 features
• Learn how to write your own primitive to be applied to the data

Running the tutorial

If you would like to work on Kaggle, the Tutorial notebook has been uploaded as a kernel. You can fork that notebook to use as a starting point. If you prefer to work locally:

1. Clone the repo

   git clone https://github.com/Featuretools/predict-household-poverty.git
   

2. Install the requirements

   pip install -r requirements.txt
   

   You will also need to install graphviz for this demo. Please install graphviz according to the instructions in the Featuretools Documentation

3. Download the data

   You can download the data from Kaggle or create a kernel and use Featuretools there. After downloading, save the CSV to a directory called data in the root of this repository.

4. Run the Tutorial notebook

   jupyter notebook
   

Feature Labs

Featuretools is an open source project created by Feature Labs. To see the other open source projects we're working on visit Feature Labs Open Source. If building impactful data science pipelines is important to you or your business, please get in touch.

Contact

Any questions can be directed to [email protected]