Quantification of misclassification categorisation. Misclassification categories quantified across all testers and clinicopathologic findings for false positives and false negatives in a pilot test of the text-mining application; DEE, designed for the rapid retrieval of necropsy data from free-text reports within the Wildbase Pathology Register of Aotearoa New Zealand.

Documents similaires

• Bar charts of false positive misclassifications for tested clinicopathologic findings across all testers.
  par: Stefan Saverimuttu (20882325)
  Publié: (2025)
• Bar charts of false negative misclassifications for tested clinicopathologic findings across all testers.
  par: Stefan Saverimuttu (20882325)
  Publié: (2025)
• Table of misclassification coding structure. Codes in the left-hand column were given to individual records misclassified by the application according to the false positive and false negative code definitions and examples described.
  par: Stefan Saverimuttu (20882325)
  Publié: (2025)
• Statistics of search terms and stems. Mean, standard deviation, and range of the number of selected search terms and number of 3 letter word stems derived from these search terms, across testers and clinicopathologic findings in the pilot test of an application for extraction of necropsy data from the Wildbase Pathology Register, DEE.
  par: Stefan Saverimuttu (20882325)
  Publié: (2025)
• Metrics of application performance. Mean, standard deviation, and range of recall, precision, and calculated F1-score of all four tested clinicopathologic findings across all testers of the application for extraction of necropsy data from the Wildbase Pathology Register, DEE.
  par: Stefan Saverimuttu (20882325)
  Publié: (2025)