jdposada / FeatureSelectionComparison

PatientLevelPrediction methods research investigating different feature selection approaches for simplifying models

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Comparison of different feature selection pipelines for creating parsimonious patient-level prediction models

Study Status: Started

  • Analytics use case(s): Patient-Level Prediction
  • Study type: Methods Research, Clinical Application
  • Tags: Feature Selection
  • Study lead: Jose Posada, Aniek Markus, Ross Williams, Jenna Reps
  • Study lead forums tag: jposada jreps RossW aniekmarkus
  • Study start date: 2021-11-15
  • Study end date: -
  • Protocol: Click Here
  • Publications: -
  • Results explorer: -

Method research to determine

  1. Whether we can implement feature selection in big p and big n data to develop parsimonious models while not compromising model performance
  2. Whether one feature selection pipeline is better than others

About

PatientLevelPrediction methods research investigating different feature selection approaches for simplifying models


Languages

Language:TeX 83.5%Language:CSS 16.5%