Constructive induction, viewed generally, is a process combining two intertwined searches: first for the best representation space, and second for the best hypothesis in that space. The first search employs operators for improving the initial representation space, such as those for generating new attributes, selecting best attributes among the given ones, and abstracting attributes. In the methodology presented, these operators are chosen on the basis of an analysis of the training data, hence the term data-driven. The second search employs an AQ inductive learning method to the examples projected at each iteration into the newly modified representation space. The aim of the second search is to determine a generalized description of examples. Experimental applications of the methodology to text categorization and natural scene interpretation demonstrate a significant practical utility of the proposed methodology.
CITATION STYLE
Bloedorn, E., & Michalski, R. S. (1998). Data-Driven Constructive Induction: Methodology and Applications. In Feature Extraction, Construction and Selection (pp. 51–68). Springer US. https://doi.org/10.1007/978-1-4615-5725-8_4
Mendeley helps you to discover research relevant for your work.