Data mining application to healthcare fraud detection: a two-step unsupervised clustering method for outlier detection with administrative databases

Table 5 Percentile of Variables Distributions each analyzed outlier falls into

Variable	H11	H31	H51
Avg Cost	0.994	0.937	0.734
Percent CC	0.399	0.899	0.981
Specialization	0.994	0.816	0
Number of Visits (patients)	0.532	0.994	0.032
Upcoding Index	0.513	0.506	0.890

ISSN: 1472-6947