人工智能课件AI_Chap01TraditionalAI1DecisionTree幻灯片.pptxVIP

Artificial Intelligence Decision TreeoutlooksunnyrainyyescompanynobigmednosailboatyessmallbigyesnoAn Example Data Set and Decision TreeClassificationoutlooksunnyrainyyescompanynobigmednosailboatyessmallbigyesnoDecision Trees A hierarchical data structure that represents data by implementing a divide and conquer strategy Given a collection of examples, learn a decision tree that represents it. Use this representation to classify new examplesBACColor BlueredGreenShapeShapeBtrianglecirclecirclesquaresquareBABCADecision Trees: The Representation Decision Trees are classifiers for instances represented as features vectors.(color= ;shape= ;label= ) Nodes are tests for feature values; There is one branch for each value of the feature Leaves specify the categories (labels) Can categorize instances into multiple disjoint categories(color= RED ;shape=triangle)Evaluation of a Decision TreeLearning a Decision TreeColor BlueredGreenShapeShape+trianglecirclecirclesquaresquare+--++Decision TreesOutlook RainSunnyOvercastHumidityWindHighNormalStrongWeakYesYesNoNoYesInduction of Decision TreesData Set (Learning Set)Each example = Attributes + ClassInduced description = Decision treeTDIDTTop Down Induction of Decision TreesRecursive PartitioningSome TDIDT SystemsID3 (Quinlan 79)CART (Brieman et al. 84)Assistant (Cestnik et al. 87)C4.5 (Quinlan 93)See5 (Quinlan 97)…Secondary Gleason Grade1,2345NoYesPSA LevelStage?14.9?14.9T1ab,T3T1c,T2a,T2b,T2cNoNoYesPrimary Gleason Grade2,34NoYesProstate cancer recurrenceTDIDT AlgorithmResulting tree T is:Attribute AAv1v2vnA’s valuesT1T2TnSubtreesLearning Algorithm DT(Examples, Attributes) If all Examples have same label: return a leaf node with Label Else If Attributes is empty: return a leaf with majority Label Else Pick an attribute A as root For each value v of A Let Examples(v) be all the examples for which A=v Add a branch out of the root for the test A=v If Examples(v) is empty create a leaf node labeled with the majority label in E