Analyzing Credit Scores with tidymodels in RKey Takeaways:
- Learn the benefits of dimensionality reduction.
- Learn to perform feature extraction tidymodels recipes.
- Learn incorporate feature extraction in the model building process using tidymodels workflows.
In this live training, we’ll motivate the benefits of dimensionality reduction while exploring predictors of credit scores. Using ggplot2, we’ll demonstrate how UMAP can extract information-rich features that help to group credit scores. Then, we’ll build UMAP into a tidymodels workflow that fits a decision tree model to predict credit scores. We’ll evaluate the performance of models with and without UMAP dimensionality reduction.
Matt is an Associate Professor of Data and Analytics at Northern Illinois University. On the side, he does data analytics consulting and training as the owner of Pickard Predictives, LLC. He’s an avid user of R, and competent in Python. His data interests and work center on NLP, machine learning, statistical modeling, and visualization. He’s learning more about deep learning.
He's happily married with four girls and a boy poodle (who doesn’t get to vote, so Matt is still quite outnumbered). In his free time, he likes to read, bake bread, woodwork, and backpack. He’s a Jack of all trades and a master of none.