UC has developed software to measure skill proficiency which can be integrated across learning environments to objectively track skill acquisition in trainees. Our software uses Bayesian knowledge tracing paired with skill decay models to track skill acquisition during training and predict decay months after training.

Training Effectiveness Measurement (TEM)

A lightweight learning analytics solution that is added as a widget to a learning management system and content authoring platform. The software module is a training content agnostic and can be used to objectively assess the efficacy of multiple trainers teaching the same content. The software comes with a unique innovative feature to reduce the need for huge amounts of data compared to traditional machine learning models. Over time, as more data from multiple cohorts is accumulated, the model updates the parameters in an unsupervised manner to suit the given student distribution. Our studies have shown that remediation based on acquired skill proficiency results in better on-the-job performance and longer skill retention (6 months post training). The software has been used with Navy maintenance trainers, VR maintenance trainers, and Air Force MOTAR digital training platform.