ActiveTigger: An open source collaborative text annotation software for computational social sciences

ActiveTigger is an open source software tool designed to support collaborative text annotation for computational social scientists. It implements in a user-friendly interface machine learning features such as training classifiers, fine-tuning language models, evaluating performance, and running predictions on large datasets. In particular, it implements active learning as an intermediate step between human annotation and machine learning. This article reviews its origins and develoments, introduces its main features, and discusses its current limitations and future directions.