Decision trees have been already successfully used in medicine, but as in t
raditional statistics, some hard real world problems can not be solved succ
essfully using the traditional way of induction. In our experiments we test
ed various methods for building univariate decision trees in order to find
the best induction strategy. On a hard real world problem of the Orthopaedi
c fracture data with 2637 cases, described by 23 attributes and a decision
with three possible values, we built decision trees with four classical app
roaches, one hybrid approach where we combined neural networks and decision
trees, and with an evolutionary approach. The results show that ail approa
ches had problems with either accuracy, sensitivity, or decision tree size.
The comparison shows that the best compromise in hard real world problem d
ecision trees building is the evolutionary approach. (C) 2001 Elsevier Scie
nce Ireland Ltd. All rights reserved.