GoFundMe's mission is to help people help each other by making it safe and easy for people to ask for help and support the causes they care about. Since 2010, GoFundMe has become a trusted leader in online fundraising, with $17 billion raised from over 200 million donations. In 2022, GoFundMe acquired Classy, the leading nonprofit fundraising software company, which operates as a wholly owned subsidiary of GoFundMe. Together, the two companies serve as a global leader in giving for individuals and nonprofits, accelerating growth and unlocking new opportunities to help more people and organizations. Our vision is to become the most helpful place in the world--join us!

Interpretable classifiers have recently witnessed an increase in attention from the data mining community because they are inherently easier to understand and explain than their more complex counterparts. Examples of interpretable classification models include decision trees, rule sets, and rule lists. Learning such models often involves optimizing hyperparameters, which typically requires substantial amounts of data and may result in relatively large models. In this paper, we consider the problem of learning compact yet accurate probabilistic rule lists for multiclass classification. Specifically, we propose a novel formalization based on probabilistic rule lists and the minimum description length (MDL) principle. This results in virtually parameter-free model selection that naturally allows to trade-off model complexity with goodness of fit, by which overfitting and the need for hyperparameter tuning are effectively avoided. Finally, we introduce the Classy algorithm, which greedily finds rule lists according to the proposed criterion. We empirically demonstrate that Classy selects small probabilistic rule lists that outperform state-of-the-art classifiers when it comes to the combination of predictive performance and interpretability. We show that Classy is insensitive to its only parameter, i.e., the candidate set, and that compression on the training set correlates with classification performance, validating our MDL-based selection criterion.

Classy started as a code challenge for a technical interview for a machine learning job in Tokyo. The challenge was to create an an application that uses a supervised learning algorithm to classify images based on their text content. This application was then wrapped in a server with a tiny REST API, and Classy was born. I can certainly make that happen! Let me know a little more about what you're working on using the contact form below and we'll get the ball rolling.