Developing news-based Economic Policy Uncertainty index with unsupervised machine learning

•Employ LDA to recover topics underpinning aggregate Economic Policy Uncertainty.•The approach economizes on costly human classification to pre-define a set of keywords.•EPU index derived using LDA is validated with the one derived using existing methods.

I propose creating a news-based Economic Policy Uncertainty (EPU) index by employing an unsupervised algorithm able to deduce the subject of each article without the need for pre-labeled data.