Identifying text discourse structure of the narratives describing psychiatric patients’ defense mechanisms

  • Authors:
  • Eunmi Ham;Woojin Paik

  • Affiliations:
  • Dept. of Nursing Science, Konkuk University, Korea;Dept. of Computer Science, Konkuk University, Korea

  • Venue:
  • FinTAL'06 Proceedings of the 5th international conference on Advances in Natural Language Processing
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

Psychiatric nursing care plans include the narratives describing the defense mechanisms exhibited by the patients. These narratives form the basis for generating psychodynamic analysis, which is one of the key diagnosis outcomes about the patients. However, it is fairly difficult for the novice nurses to correctly identify the type of defense mechanism based on the observations that they made while caring for the patients. One of the main reasons for the high error rate is the lack of uniform terminology. That is, inconsistencies in the definitions and conceptualizations of defenses. Furthermore, there is lack of sufficient examples showing the wide variety of cases from which the novice nurses to learn. We developed a prototype text discourse analysis system, which assigns one or more text discourse categories to each clause in the defense mechanism narratives. The initial evaluation of the prototype system resulted in correctly identifying 85% of the defense mechanisms in the test data set. The output from the text discourse analysis system is fed into a database to augment the definition of the defense mechanisms and also to be used as a learning tool for the novice nurses.