University uses machine learning on ID cards to predict dropouts

14/03/18

A new ID card tracking system being used by the University of Arizona keeps a record of how often students interact in social settings on campus (like how frequently they use the campus rec center), what they buy to eat, and their academic performance.

According to the University, the data allows them to usually predict (quite accurately) within a freshman’s first 4 weeks if they will return as a sophomore and eventually graduate.

Based on the data, the university creates a list every quarter of freshman in danger of dropping out and shares it with the students’ advisors, who do their best to intervene.

So far, their efforts have been pretty successful. After 3 years of collecting freshman data, their predictions have been 73% accurate. And, in 2017 the school’s retention rate rose to 86.5% — about 11% higher than the national average. 

The social data the school gathers includes timestamps and locations, and according to Gizmodo, the university never discloses how card swipes and payments are used to monitor student behavior on their policy site. 

Subscribe to our free newsletter
Follow us on Twitter
Join us on LinkedIn

Latest Features & Interviews

White paper: Gemalto - The paper elements of a passport

An expert’s guide to using the paper elements of a passport.

INTERVIEW: Gemalto‘s Xavier Larduinat on the future of borders and digital identities

Xavier Larduinat, Marketing & Communication manager for Innovation at Gemalto, explores the future of digital identity, the Thales acquisition and other trends in an exclusive interview with Security Document World.

Interview: Sean Farrell, Head of Portfolio Management, Government Solutions, SITA

Planet Biometrics discussed the future of biometric travel technology with Sean Farrell, head of Portfolio Management, Government Solutions, SITA.

More articles >>
Share |

Sponsored Links

SDW Conference and Exhibition
SDW is a world-leading conference and exhibition providing a global showcase for next-generation secure credentialing solutions.