Improving Emergency Data Quality by Noisy and Error Identification

Authors

  • Gong Zhang
  • Trevor Strome
  • Simon Liao
  • Zhaopeng Fan

Abstract

For emergency department data, it is important to achieve the high quality data. Biomedical data errors or noise tend to fall in data entry inaccuracy, medical device limitations, data transmission errors, or man-made perturbations frequently result in imprecise or vague data. To find these errors or noise, a data mining algorithm was build to for identifying errors or noisy values and using the remaining correct data sets for subsequent modeling and analysis. This approach was to build a business rule based naïve bayes classifier. With this model, the error or noisy patterns can be discovered from datasets that were feeding to emergency room data repository.

Downloads

Published

2010-06-15

How to Cite

[1]
G. Zhang, T. Strome, S. Liao, and Z. Fan, “Improving Emergency Data Quality by Noisy and Error Identification”, CMBES Proc., vol. 33, no. 1, Jun. 2010.

Issue

Section

Posters