|
| | | |
We wish you a happy and safe holiday season and all the best for 2025
Discovering inappropriate billings with local density based outlier detection method
Shan, Y., Murray, D. W. and Sutinen, A.
This paper presents an application of a local density based outlier detection method in compliance in the context of public health service management. Public health systems have consumed a significant portion of many governments� expenditure. Thus, it is important to ensure the money is spent appropriately. In this research, we studied the potentials of applying an outlier detection method to medical specialist groups to discover inappropriate billings. The results were validated by specialist compliance history and direct domain expert evaluation. It shows that the local density based outlier detection method significantly outperforms basic benchmarking method and is at least comparable, in term of performance, to a domain knowledge based method. The results suggest that the density based outlier detection method is an effective method of identifying inappropriate billing patterns and therefore is a valuable tool in monitoring medical practitioner billing compliance in the provision of health services. |
Cite as: Shan, Y., Murray, D. W. and Sutinen, A. (2009). Discovering inappropriate billings with local density based outlier detection method. In Proc. Australasian Data Mining Conference (AusDM'09) Melbourne, Australia. CRPIT, 101. Kennedy P. J., Ong K. and Christen P. Eds., ACS. 93-98 |
(from crpit.com)
(local if available)
|
|