Inductive algorithms rely strongly on their representational biases. Representational inadequacy can be mitigated by constructive induction. This chapter introduces several important issues for constructive induction and describes a new multi-strategy constructive induction algorithm (GALA2.0) which is independent of the learning algorithm. Unlike most previous research on constructive induction, our methods are designed as a preprocessing step before standard machine learning algorithms are applied. We present the results which demonstrate the effectiveness of GALA2.0 on real domains for two different learners: C4.5 and backpropagation.
|Title of host publication||Feature Extraction, Construction and Selection|
|Place of Publication||Boston|
|Number of pages||16|
|State||Published - 1998|
|Name||The Springer International Series in Engineering and Computer Science|