作者: James Geller , None
DOI: 10.1016/B978-0-444-81704-4.50011-9
关键词:
摘要: This chapter elaborates on an approach to knowledge representation that combines the use of a limited inference strategy with massive parallelism. Conceptually, class hierarchy is used. The nodes in this are augmented by preorder numbering scheme. then transformed into pointer-free linear tree . implemented CM-2 Connection Machine, such every node resides its own processor. discusses detail algorithm for inheritance representation. It also introduces upward-inductive Experimental data from Machine implementation show given machine size downward can be performed constant time. height has no influence run Upward-inductive times grow very moderately number tree. 1