Heather Crombie

Heather Crombie Heather Crombie Heather Crombie
  • Home
  • Career
    • Resume
    • Strategic Research
    • Process
    • Case Studies
  • More
    • Home
    • Career
      • Resume
      • Strategic Research
      • Process
      • Case Studies

Heather Crombie

Heather Crombie Heather Crombie Heather Crombie
  • Home
  • Career
    • Resume
    • Strategic Research
    • Process
    • Case Studies

Training Support for Machine Learning

Working closely with our Quant & User Support teams a researcher hand  coded over 5,000 open text responses submitted in our application from our users. These were used to train  a machine algorithm which was built and is now used to identify user feedback patterns. The issues identified are then assigned to the relevant team to fix or redesign. 

Actions & Outcomes from Machine Learning

Initial Conclusions - Spring 22

Feedback primarily concerned challenges 

  • Difficulty locating assignments
  • Multiple class Connects Assigned simultaneously
  • Broken compute issues
  • Too much work
  • Too many class connect sessions
  • Expectation that the program would be more flexible
  • Difficulties with multiple students

Subsequent Analysis - Summer 22

Rapid analysis means more insights, so far the Algorithm has been used for the following analyses:
 

  • Connection speeds by region
  • Class Connect issues by school
  • Withdrawal focus group transcripts
  • Market research data
  • Data extracted for IOL courses
  • Targeted programming developed by
    School Services


Copyright © 2024 Heather Crombie Portfolio of Work - All Rights Reserved.

  • Resume

This website uses cookies.

We use cookies to analyze website traffic and optimize your website experience. By accepting our use of cookies, your data will be aggregated with all other user data.

Accept