Welcome to Dataro’s Tech Explainer series where our CTO, Dave Lyndon, talks through some of the most commonly asked questions in fundraising tech.
After our first tech user group, there were many questions about what features are used by Dataro’s Machine Learning (ML) models.
In order to properly explore this question, it’s important to understand how ML models make use operate of their input features. We think a good way to understand this is by working with three features that many fundraisers already use in their own RFM segmentation. We will train some models with these features, observe how the data is prepared and how the models are evaluated. In addition, we will explore what the models have learned and cover some basics of the learning algorithms.
In the first part of the series, we examined RFM from the ML perspective. In part two, we dove into learning algorithms with some linear models: Linear Regression & Logistic Regression.
In part three, we look at a more sophisticated class of models: tree-based models (Decision Trees, Random Forest & Boosted Trees). These models have numerous advantages to linear models in the fundraising context. Again there will be some math and code, but the aim is to keep everything as approachable as possible!
We’re looking forward to seeing you there!