Course

Course Summary
Credit Type:
Course
ACE ID:
STAT-0044
Organization's ID:
668
Organization:
Length:
4 weeks (60 hours)
Dates Offered:
Credit Recommendation & Competencies
Level Credits (SH) Subject
Graduate 3 data analytics
Description

Objective:

The course objective is to cover the application of predictive modeling algorithms to healthcare scenarios and data. You will learn how to formulate predictive modeling problems, train different types of predictive models on data where the target outcome is known, assess the performance of those models using holdout data not used in training, and use the resulting models to predict new data. You will also learn how to audit models for potential bias and unfairness.

Learning Outcomes:

  • explain the structure of medical data and the challenges this poses
  • specify the predictive modeling task
  • fit machine learning models with linear and logistic regression, and decision trees using R Assess the performance of these models with holdout data
  • choose and modify models to improve performance
  • generate predictions for new data using the models
  • audit models for potential bias and unfairness

General Topics:

  • The predictive modeling approach
  • Regression models
  • Decision trees and neural nets
  • Concluding topics and project
Instruction & Assessment

Instructional Strategies:

  • Computer Based Training
  • Discussion
  • Practical Exercises

Methods of Assessment:

  • Quizzes
  • Project

Minimum Passing Score:

80%
Supplemental Materials