Bayesian Classification

Transcript

1. Bayesian Classifier Gang Tao

2. Algebraic Geometry Complex Analysis factal Differential equation Geometry Dynamical System

   Combinatorial Mathematics Statistics Computational mathematics

3. Bayes Theorem

4. None

5. Bayes Theorem

6. Diachronic Interpretation H -> Hypothesis D -> Data P(H) ->

   Prior Probability P(H|D) -> Posterior Probability P(D|H) -> Likelihood P(D) -> Normalizing Constant

7. Bayes Theorem Original Belief Observation + = New Belief

8. Bayes and Occam’s Razor

9. “All Models are wrong, but some of them are better

   than the others”

10. Model Complexity

11. Naive Bayes “Naive” because it is based on independence assumption

    All the attributes are conditional independent given the class

12. Naive Bayes Classifier

13. How to build a Bayesian Classifier for prediction Prepare Data

    Features Extraction Select Distribution Model Calculate the Probability for each attributes Multiply All Probabilities Label with highest Probability

14. Advantage VS. Disadvantage Powerful Efficient in Space and Time Incremental

    Trainer Simple Independant Assumption Probability are not relevant

15. Application of Bayesian Classifier Spam Email Filter Natural Language Processing

    Word Segmentation Spell Checking Machine Translation Pattern Recognition

16. Thank You