Theory Formation in the Decision Trees Domain

概要

論文の詳細を見る
Decision trees are widely used in machine learning and knowledge acquisition systems. However, there is no optimal or even unanimously accepted strategy of obtaining "good" such trees, and most of the generated trees suffer from inadequacies in representing knowledge. The final goal of our research is to formulate a theory for the decision trees domain, that is a set of heuristics (on which a majority of experts will agree) which will describe a good decision tree and will define a measure of the quality of decision trees, as well as a set of heuristics specifying how to obtain optimal trees. In order to achieve this goal we have designed a recursive architecture learning system, which monitors an interactive knowledge acquisition system using decision trees, and incrementally acquires from the experts using it the knowledge used to build the decision tree domain theory. Our system interactively acquires knowledge to define the notion of good/bad decision trees and to measure their quality, as well as knowledge needed to specify ways of constructing good decision trees. The partial theory acquired at each moment is also used by the basic knowledge acquisition system in its tree generation process, thus constantly improving its performance.
社団法人人工知能学会の論文
1992-05-01