We found substantial annotator variability in identifying supporting, refuting, or uncertain evidence for the diagnosis of pneumonia in clinical text. Future work will expand these methods to a larger case sample and incorporate a more formal linguistic analysis to identify specific lexical cues thereby extending existing taxonomies of uncertainty and improving automated NLP algorithms.
Authors own copyright of their articles appearing in the Online Journal of Public Health Informatics. Readers may copy articles without permission of the copyright owner(s), as long as the author and OJPHI are acknowledged in the copy and the copy is used for educational, not-for-profit purposes. Share-alike: when posting copies or adaptations of the work, release the work under the same license as the original. For any other use of articles, please contact the copyright owner. The journal/publisher is not responsible for subsequent uses of the work, including uses infringing the above license. It is the author's responsibility to bring an infringement action if so desired by the author.