Figure 1 shows a sample data including the two classes. Both logistic regression and Naive Bayes methods can be applied to classify the data without any error. One reason is that there is a gap area known as a decision boundary that logistic regression and Naive Bayes can search for to separate the two classes. Figure 2 illustrates a decision boundary that can be used for classification.

Figure 1

enter image description here

Figure 2

enter image description here Finding a decision boundary can difficult in some other dataset. If logistic regression or Naive Bayes cannot find a correct decision boundary, their classification results would involve a serious error. Can you provide an example of 2-dimentaional data in the format of Figure 2 for each of the following situations?

1) Logistic regression can find a decision boundary correctly, whereas Naive Bayes cannot.

2) Naive Bayes can find a decision boundary correctly, whereas logistic regression cannot.

  • $\begingroup$ You are assuming the boundary is real. See this. $\endgroup$ – Frank Harrell May 20 '18 at 11:47

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.