Why Is Rule Learning Optimistic and How to Correct It

@inproceedings{Mozina2006WhyIR,
  title={Why Is Rule Learning Optimistic and How to Correct It},
  author={Martin Mozina and Janez Demsar and Jure Zabkar and Ivan Bratko},
  booktitle={ECML},
  year={2006}
}
In their search through a huge space of possible hypotheses, rule induction algorithms compare estimations of qualities of a large number of rules to find the one that appears to be best. This mechanism can easily find random patterns in the data which will – even though the estimating method itself may be unbiased (such as relative frequency) – have optimistically high quality estimates. It is generally believed that the problem, which eventually leads to overfitting, can be alleviated by… CONTINUE READING