Alexander Ratzan

Posts by Collection

portfolio

projects

All-NBA Machine Learning

Predicted the 2020 All-NBA team with 87% accuracy using a multi-layer perceptron neural net optimized and trained on 20 seasons of NBA statistics.
Figures: Left - model correctness for 2020 All-NBA team. Right - feature weights for center position (based on a comparable accuracy random forest model trained on same dataset)


NeuroAnalysis Project

Applied various machine learning techniques (Support Vector Machines, Clustering, Multi-Dimensional
Scaling, Principal Component Analysis) and statistical analysis tools (Representational Similarity Analysis, correlational maps and matrices) to classify, predict, and examine neural data.


Figures: Left - seed region where fMRI data was extracted from when viewing an item. Middle - 2D representation of data from PCA. Right - 3D representation from PCA.

Clinical Subtyping of Multiple Sclerosis with SuStaIn

Here, I led a project at Columbia Medical Center to generate informative clinical subtypes in a large dataset of patients with MS. In evaluating the clinical features available, I chose to apply a self-optimized version of SuStaIn (sub typing and staging inference) and identified 3 subtypes that differed in their disease progression as well as demographic and neurophysiological features.

Figures: Left - kernel density estimates of clinical features between patients and asymptomatic patients. Right - derived subtypes from patient data.

publications

talks

teaching

Teaching experience 1

Undergraduate course, University 1, Department, 2014

This is a description of a teaching experience. You can use markdown like any other post.

Teaching experience 2

Workshop, University 1, Department, 2015

This is a description of a teaching experience. You can use markdown like any other post.