表10-1学习析取规则集的序列 覆盖算法(CN2) Sequential-Covering(Target attribute, Attributes, Examples Threshold) Learned rulest RuletLearn-One-Rule Target attribute, Attributes Examples Performance(Rule, Examples)>Threshold Learned rules< Learned rules+ rule Examples∈ Examples-{被Rule正确分类的样例} Rulet-Learn-One- Rule(Target attribute, Attributes, Examples) Learned rules∈按照在 Examples上的 Performance排序的 earned rules 返 回 Learned rules 2003.12.18机器学习-学习规则集合作者: Mitchell译者:曾华军等讲者:陶晓鹏2003.12.18 机器学习-学习规则集合作者：Mitchell 译者：曾华军等讲者：陶晓鹏 8 表10-1 学习析取规则集的序列 覆盖算法（CN2） Sequential-Covering(Target_attribute, Attributes, Examples, Threshold) • Learned_rules{} • RuleLearn-One-Rule(Target_attribute, Attributes, Examples) • 当Performance(Rule, Examples)>Threshold – Learned_rulesLearned_rules+Rule – ExamplesExamples-{被Rule正确分类的样例} – RuleLearn-One-Rule(Target_attribute, Attributes, Examples) • Learned_rules按照在Examples上的Performance排序的 Learned_rules • 返回Learned_rules