Abstract: | It is recognised that the problem space for constructive induction is usually intractably large. However, the size of the space differs depending on a number of factors. The paper considers a variety of contexts and demonstrates that the problem space expands dramatically in the case where iterative feature construction is used. However, it argues that, even though the complexity of iterative construction is extreme, there are situations in which it is a necessary part of learning. An illustrative example is provided.
|