graphicalModels_4perPage - Outline Graphical Models Data...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Graphical Models Data Mining Prof. Dawn Woodard School of ORIE Cornell University 1 Outline 1 Graphical Models 2 Naive Bayes Assumption: Pr ( { X 1 , X 2 ,..., X K }| Y )= K Y k = 1 Pr ( X k | Y ) Naive Bayes assumes that, conditional on the outcome variable, the predictors are independent. 4 Credit Variables Recall the variables for the credit data: Credit V1 V14 V2 V3 5
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
A Full Model A full model for Pr ( Y , { X k } ) must take into account all the two-way, three-way, etc. interactions between the variables: Credit V1 V14 V2 V3 6 Naive Bayes Naive Bayes only models the interaction between the outcome and each of the predictors: Credit V1 V14 V2 V3 7 Naive Bayes This type of graph is called a conditional independence graph , graphical model , Bayesian network ,or belief network , and shows the model’s independence assumptions: Credit V1 V14 V2 V3 8 Credit Data Do you think the naive Bayes assumption holds (approximately) for the credit data? Recall that the predictors are variables like: Amount in checking account Duration of credit in months Credit history Duration of present employment Marital status 9
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 5

graphicalModels_4perPage - Outline Graphical Models Data...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online