INDUCTIVE LEARNING FROM PRECLASSIFIED TRAINING EXAMPLES - AN EMPIRICAL-STUDY

Many real-world decision-making problems fall into the general categor y of classification. Algorithms for constructing knowledge by inductiv e inference from example have been widely used for some decades. Altho ugh these learning algorithms frequently address the same problem of l earning from preclassified examples and much previous work in inductiv e learning has focused on the algorithms' predictive accuracy, little attention has been paid to the effect of data factors on the performan ce of a learning system. An experiment was conducted using five learni ng algorithms on two data sets to investigate how the change in labeli ng the class attribute can alter the behavior of learning algorithms. The results show that different preclassification rules applied on the training examples can affect either the classification accuracy or cl assification structure.