第六章 学习规则集合.pptVIP

Chapter 10 学习规则集合Learning Sets Of Rules Tom M.Mitchell Carnegie Mellon University Content Introduction Sequential Covering Algorithm Learning First-Order Rules (FOIL Algorithm) Induction As Inverted Deduction（作为逆演绎的归纳） Inverting Resolution（逆归纳） Introduction Representation of learned hypothesis — sets of if-then rules(命题规则集) Learning Disjunctive Sets of Rules Method 1: Learn decision tree convert to rules Method 2: Sequential covering algorithm(序列覆盖) Learn sets of first-order rules that contain variables Learn one rule at a time Introduction Introduction Representation of learned hypothesis — sets of if-then rules Learning Disjunctive Sets of Rules Method 1: Learn decision tree convert to rules Method 2: Sequential covering algorithm Learn sets of first-order rules that contain variables Learn one rule at a time Introduction An example of first-order rule sets target concept: Ancestor IF Parent(x,y) THEN Ancestor(x,y) IF Parent(x,y)∧Ancestor(z,y) (conjuctive) THEN Ancestor(x,y) The content of this chapter Learning algorithms capable of learing such rules , given sets of training examples Content Introduction Sequential Covering Algorithm Learning First-Order Rules (FOIL Algorithm) Induction As Inverted Deduction Inverting Resolution Sequential Covering Algorithm Strategy of this algorithm Learn one rule with high accuracy any coverage Remove positive examples covered by this rule Repeat Sequential Covering Algorithm Learn-One-Rule Sequential Covering Algorithm Sequential Covering Algorithm .vs. Other Algorithm ID3 –simultaneous covering algorithm the number of distinct choices to learn the same set of rules Numsequential NumID3 (搜索次数) The Direction of the Search Sequential : General to Specific (逐项增加属性) Find-s : Specific to General(以最特殊的实例开始) Sequential : genetrate then test search Find-s ,Candidate-Elimination: example-driven search Rules