Automated Free Text Analysis Methodologies for the Total Army Injury and Health Outcomes Database.

摘要

This project investigated how data mining techniques could be applied to the Total Army Injury and Health Outcomes Database (TAIHOD), developed by the U.S. Army Research Institute for Environmental Medicine (USARIEM). In particular, we explored the potential for 'information mining', a unique fusion of techniques from data mining with methods for automatically extracting data from the unstructured text reports that often accompany medical records. Data extracted from text may provide useful elements that could contribute to the predictive accuracy of statistical models constructed with data mining techniques.