Factuality Detection using Machine Translation -- a Use Case for German Clinical Text
This work addresses data scarcity for factuality detection in German clinical text, but it is incremental as it applies an existing translation method to a specific domain.
The paper tackled the problem of factuality detection in German clinical text by using machine translation to convert English data to German, enabling the training of a transformer-based model without sharing sensitive clinical data.
Factuality can play an important role when automatically processing clinical text, as it makes a difference if particular symptoms are explicitly not present, possibly present, not mentioned, or affirmed. In most cases, a sufficient number of examples is necessary to handle such phenomena in a supervised machine learning setting. However, as clinical text might contain sensitive information, data cannot be easily shared. In the context of factuality detection, this work presents a simple solution using machine translation to translate English data to German to train a transformer-based factuality detection model.