Decision rules

From Wikipedia, the free encyclopedia

A set of decision rules is the verbal equivalent of a graphical decision tree, which specifies class membership based on a hierarchical sequence of (contingent) decisions. Each rule in a set of decision rules therefore generally takes the form of a Horn clause wherein class membership is implied by a conjunction of contingent observations.

IF condition_{1} AND condition_{2} AND ... AND condition_{n} THEN CLASS = class_{i}

where condition_{j} is in general contingent on the choice of condition_{{j-1}}. Decision rules can be transcribed from the corresponding decision tree, or can be induced directly from observations.

Decision rules are commonly used in the medical field. For example, the Ottawa Ankle Rules guide obtaining radiographs for traumatic ankle pain.

C4.5 is a well-known tool for discovering decision rules. There has been work done on discovering temporal decision rules, where the condition attributes are observed before the decision attribute. [1]


References

  1. ↑ Generation and Interpretation of Temporal Decision Rules, Kamran Karimi and Howard J. Hamilton, International Journal of Computer Information Systems and Industrial Management Applications (IJCISIM), Volume 3, 2011
This article is issued from Wikipedia. The text is available under the Creative Commons Attribution/Share Alike; additional terms may apply for the media files.