IMPROVING PREDICTION OF PRETERM BIRTH USING A NEW CLASSIFICATION SCHEME AND RULE INDUCTION

Prediction of preterm birth is a poorly understood domain. The existin g manual methods of assessment of preterm birth are 17% - 38% accurate . The machine learning system LERS was used for three different datase ts about pregnant women. Rules induced by LERS were used in conjunctio n with a classification scheme of LERS, based on ''bucket brigade algo rithm'' of genetic algorithms and enhanced by partial matching. The re sulting prediction of preterm birth in new, unseen cases is much more accurate (68%-90%).