#frAIday: Transformer-based variables for estimating hate crimes in police reports
Fri
20
Sep
Friday 20 September, 2024at 12:15 - 13:00
Galaxen & Zoom
To address the challenge of estimating population totals in large textual datasets within official statistics, where manual annotation is impractical, we propose a method combining transformer encoder neural network predictions and well-established survey sampling estimators. This is done by training a classifier and then using the model predictions as an auxiliary variable in the estimators. The applicability is demonstrated on Swedish hate crime statistics, which are based on Swedish police reports, for which approximately 1.5 million are being filed annually.
This website uses cookies which are stored in your browser. Some cookies are necessary for the page to work properly and others are selectable. You choose which ones you want to allow.