Algorithmic Impact Assessment - Employment Insurance (E.I) Backlog Reduction (English, PDF)

View metadata and access the original English PDF file: Algorithmic Impact Assessment - Employment Insurance (E.I) Backlog Reduction.

Format

PDF

Access

file-only

Size

62,375 bytes

Publisher-reported update

2023-06-16

UPDATE: In July 2023, the implementation of the Machine Learning model proved successful, resulting in the completion of over 40,000 claims. Although this page won't receive further updates, the project is still ongoing and can be accessed here. It's now referred to as the “Employment Insurance Machine Learning Workload”. The COVID 19 pandemic has triggered an unprecedented volume of Employment Insurance (EI) claims and associated Claim Review Work. During the pandemic, the focus on implementing the Emergency Response Benefit (ERB), Simplified EI, and the subsequent return to regular EI has resulted in a backlog of claim reviews, many of them for claims established before March 2020, which are competing for resources with more current and pressing work. The implementation of the Pre-ERB EI Recalculation Outcome Prediction Machine Learning (ML) model seeks to minimize the number of older claims (pre-March 2020) requiring review by an officer by using the model to predict the most probable outcome of each recalculation and triaging the associated work items accordingly, with recalculations which are unlikely to impact the claimants. The model has been developed by Employment…

Access original file