Skip to content

Latest commit

 

History

History
20 lines (17 loc) · 911 Bytes

proposal.md

File metadata and controls

20 lines (17 loc) · 911 Bytes

Health Care Classification Proposal

Objective

Health care insurance can be found in both the public and private marketplace and due to financial constraints some Americans are uninsured as well. Lack of health insurance results in worse preventive care and services for major health conditions and chronic diseases. By using a classification model on a sample US population from 2017 I propose to analyze which demographic characteristics result in Americans having public, private or no health insurance.

Data

  • American Community Surveys (ACS)
  • ~3M observations
  • Labels: Health Insurance, Private/Public Health Insurance Coverage
  • Features: Age, Sex, Income, Race, Marital Status, Education, Veteran Status, State of Work & more

Model

  • scikit-learn
  • Classification: KNN, Logistic Regression & more

App

  • Flask
  • PostgreSQL
  • AWS
  • Visualization
  • Form