Industry
Education
Technology stack
Predictive Analysis
Machine Learning
Industry
Education
Technology stack
Predictive Analysis
Machine Learning
Industry
Education
Technology stack
Predictive Analysis
Machine Learning
The challenge
The university wanted to be able to find a way to identify students who were at risk of mental health distress in order to provide students with appropriate guidance and support to help combat student drop outs.
The solution
An application of the Early Warning System was used to detect early signs of mental health distress in students.
Feature engineering was used to create a student vector that combines demographic data, past and current enrolment information and activity recorded online such as Blackboard, WiFi, library and edX activity. Using the student vector, a classification model was trained and stored for inference.Â
The outcomes
The model created by Blackbook is capable of flagging students that are potentially experiencing mental health issues.
From this project, we found that predicting outcomes based on human behaviour cannot be approached with a one-size-fits-all approach. It is the deviation from the norm at the individual level that is important, and using early warning systems powered by machine learning can efficiently uncover these deviations in a timely manner.
