Fuzzy ARTMAP Neural Network Compared to Linear Discriminant Analysis Prediction of the Length of Hospital Stay in Patients with Pneumonia

c a r e d a t a b a s e s , p o t e n t i a l l y o v e r c o m i n g o b s t a c l e s ar is ing from the n u m b e r of cases , miss ing data , v a r i a b l e s e l e c t i o n , m u l t i c o l l i n e a r i t y , s p e c i f i c a t i o n o f i m p o r t a n t i n t e r a c t i o n s , a n d s e n s i t i v i t y t o erroneous values. On a n actual database derived f rom p a t i e n t s h o s p i t a l i z e d wi th p n e u m o n i a , we compared the cross-val idated predict ions of l inear d iscr iminant analys is (LDA) t o a new, supervised adaptive resonance theory network called ARTMAP. Unbiased proport ionate reduct ion in e r r o r us ing A R T M A P w a s 50% g r e a t e r than L D A . Under c o n d i t i o n s o f s i m u l a t e d n o i s e a n d i n c r e a s i n g proportion learning, A R T M A P demonstrated further advantages o v e r LDA. The promising performance of A R T M A P warrants further evaluation on larger health care databases.