Skip to content

fsi-hack4autism/ML-To-Transform-Learner-Outcomes

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Use Case 3: Using ML To Transform Learner Outcomes

A brief description of what this project does and who it's for

Description of Files

  1. joinedData_assessmentStudentAssessor3_clean Anonymized.zip
  • Raw data for the ML model
  • Columns for the data
    • Anonymized data - StudentId, AssessmentId, OrganizationId, AssessorId
    • Assessment data - A1 through Z28. Different skill areas of Basic Learner Skills, specified in ABLLS-R Guide 2018 pg 30-49
    • Assessment metadata - FirstAssessment_byStudent, Col UC through UV
  1. ABLLS-R Guide 2018.zip
  • Provide details about the scoring model
  • Case studies to provide examples to show scoring for sampel assessments
  • The purpose of this model is to
    • identify critical skills that are in need of intervention
    • provide a method for identifying a child's specific skill
    • provide a curriculum guide for an educational program for a child
  1. Normative_Report_Hackathon23.pdf
  • This was a study done with 53 students whose data is provided in the two files below
  • examplesNeuroTypical_ABLLSR.csv
    • Similar data as in the anonymized data set, but from teh study above
    • These individual assessment scores could help:
      • offer a training data set
      • characterize skill acquisition pathways
      • give timelines in skill acquisition pathways
  • normativeBenchmarkAsessments.csv
    • This provides average assessment for each category for ages 2 years, 3 years, 4 years, and 5 years
    • There is not enough data for normalized values for 1 year patients
    • Assessment score of 0-4 is normalized from 0-1 (dividing by 4)

Sample Problem Statements

Some sample problem statements are provided below that you can adopt as the objective of your ML model. However, you are free to create a new problem statement that you believe the data in joinedData_assessmentStudentAssessor3_clean Anonymized.csv can solve

  1. Problem Statement #1 - Identify peer group for a patient to compare progress against and identify the most important features in the dataset
  2. Problem Statement #2 - Determine whether progress is being made based on assessment
  3. Problem Statement #3 - Identify if the patient is in the right program Hard as dataset does not have program information.

Live Web Page

https://jolly-desert-01b5c5610.3.azurestaticapps.net/

Team Summary

  • Dhruv BHatnagar, Presentation
  • Rishi Bhatnagar, Presentation
  • Noha Elprince, Machine Learning Expert
  • Olive Kingsley, Data Scientist
  • Anita Mohanani, QA learning AI
  • Ryan Carpenter, Jack of all trades!
  • Adam Pryce, DevOps, and Web Stack
  • Halaa Menasy, Tech Lead
  • Peter Shand, Tech Lead
  • Hanen Bondka, Shyam Sundar, Parvesh Mamgain - Data Munging

Zoom room

Join Zoom Meeting https://morganstanley.zoom.us/j/5293623419

One tap mobile +13462487799,,5293623419# US (Houston) +16694449171,,5293623419# US

About

Using ML To Transform Learner Outcomes

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages