Name Hair Height Weight Lotion

This paper addresses the issues involved in the automatic generation and upkeep of knowledge derived from datasets. Once identiication trees are generated from datasets, algorithms for converting these identiication trees into nonmonotonic inheritance structures (i.e. multiple inheritance structures with exceptions) can be applied. This results in a knowledge base which can deal with subsequent, connicting information directly without requiring a recomputation of a new identiication tree which must take the connicting information into account.