Hospital Event Reports

During my time working in the Clemson University IBM Watson in the Watt creative inquiry program, I researched the training of sentence-transformers models and BERT-style models (BERT, RoBERTa, DeBERTa) on tasks including clustering, sentiment analysis, and named entity recognition on hospital reports provided by nurses, doctors, and other hospital staff. These reports, created primarily in late 2020, consisted of a significant amount of text providing administrators feedback regarding working conditions, equipment availability, staffing, and other concerns. By developing and researching automated deep learning solutions for analyzing these reports, we provided administrators with valuable insights into overall staff feedback and key areas of concern.