Agni's Projects
Personal Website
Code for the a Dynamic Treatment Regime Framework with Trees
Installation and implementation guidelines of ICOT, a Julia-based interpretable clustering algorithm.
Repository for the ML4Business Elective for the SBS MBA program.
Available code base for the investigation: "Novel Machine Learning Provides Evidence that Stroke Risk is Not Linear: The Non-Linear Framingham Stroke Risk Score"
# StrokeDetectNLP A comprehensive framework studying the performance of simple and complex stroke-specific Natural Language Processing (NLP) and Machine Learning (ML) methods to determine presence, location, and acuity of ischemic stroke from radiographic text. ##Folders Included * Classifier: code for training across all outcomes and methods. * DataFromStructuredFiles: exploratory analysis code from structured patient data. * GloveTraining: main outputs and code related to the creation of neurology specific embeddings. We directly used [this code](https://github.com/stanfordnlp/GloVe) for training the algorithm. * Ngrams_WordCounts: exploratory analysis with n-grams and word counts. * SensitivityAnalysis: our script for the sensitivity analysis we did on our results.