1 / 40

CLASSIFICATION BOUNDARIES

CLASSIFICATION BOUNDARIES. 姓名 : 吳思葦 學號 :601630402 課堂教授 : 魏世杰 報告日期 :102/12/3 1. Weka’s Boundary Visualizer for OneR. Open iris.2D.arff , a 2D dataset Weka GUI Chooser : Visualization > BoundaryVisualizer Open iris.2D.arff Note: petallength on X , petalwidth on Y

solada
Download Presentation

CLASSIFICATION BOUNDARIES

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. CLASSIFICATION BOUNDARIES 姓名:吳思葦 學號:601630402 課堂教授:魏世杰 報告日期:102/12/31

  2. Weka’s Boundary Visualizer for OneR • Open iris.2D.arff , a 2D dataset • Weka GUI Chooser : Visualization >BoundaryVisualizer • Open iris.2D.arff • Note: petallength on X , petalwidth on Y • Choose rules > OneR • Check Plot training data • Click Start • In the Explorer , examine OneR’s rule

  3. Visualize boundaries for other schemes • Choose lazy>IBK • Plot training data ; click Start • K=5,20;note mixed colors

  4. Visualize boundaries for other schemes • Choose bayes > NaiveBayes • Set useSupervisedDiscretization to true

  5. Visualize boundaries for other schemes • Choose trees > J48 • Relate the plot to Explorer output

  6. CLASSIFICATION BOUNDARIES • Classifiers create boundaries in instance space • Different classifiers have different biases • Looked at OneR , IBK , NaiveBayes , J48 • Visualization restricted to numeric attributes , and 2D plots

  7. PREPROCESSING AND PARAMETER TUNING

  8. PREPROCESSING • Explorer可提供的資料預處理項目: • 離散化(Discretization) • 正規化(normalization) • 重新抽樣(resampling) • 屬性選擇(attribute selection) • 屬性轉換或合併(transforming and combining attributes)…

  9. Discretization • Unsupervised • weka.filters.unsupervised.attribute.Discretize • equal-width (the default) • equal-frequency • Supervised • weka.filters.supervised.attribute.Discretize

  10. Attribute Selection • 屬性評估器 • 屬性子集評估器 • 單一屬性評估器 • 搜索方法 • 搜索方法 • 排序方法

  11. Attribute Selection • 屬性評估器 • 屬性子集評估器 • 單一屬性評估器 • 搜索方法 • 搜索方法 • 排序方法

  12. Attribute Selection • 兩種屬性子集選取模式 • 屬性子集評估器+搜索方法 • 單一屬性評估器+排序方法

  13. 屬性子集評估器+搜索方法 搜索方法 BestFirst ExhaustiveSearch GeneticSearch GreedyStepwise RandomSearch RankSearch • 屬性子集評估器 • CfsSubsetEval • ClassifierSubsetEval • ConsistencySubsetEval • WrapperSubsetEval

  14. 單一屬性評估器+排序方法 • 單一屬性評估器 • ChiSquaredAttributeEval • GainRationAttributeEval • InfoGainAttributeEval • OneRAttributeEval • PrincipleComponents • ReliefAttributeEval • SymmetricalUncertAttributeEval • 排序方法 • Ranker

More Related