| Line 17: | Line 17: | ||
Overall, the notes are very good and the explanation clear and easy to follow. A few comments are written below: | Overall, the notes are very good and the explanation clear and easy to follow. A few comments are written below: | ||
| − | - At the end of section 3 "Parameter Estimation" it should be p(ω2)=Νω2/Ν | + | - At the end of section 3 "Parameter Estimation" there is a typo, it should be p(ω2)=Νω2/Ν |
| − | - | + | - In the 1D example, it is not clear how to find the separation boundary. For example, it should be mentioned that training data are used to estimate the distribution and the priors and then by equating for the two classes the boundary can be determined. Subsequently, the testing data are used to calculate the misclassification rate. |
| + | |||
| + | - Although the misclassification result (error rate) is provided in the 2D example, it is missing from the 1D example. | ||
| + | |||
| + | - | ||
---- | ---- | ||
Back to [[662slecture_tang| Bayes rule in practice: definition and parameter estimation]] | Back to [[662slecture_tang| Bayes rule in practice: definition and parameter estimation]] | ||
Revision as of 13:59, 6 May 2014
Questions and Comments for: Bayes rule in practice: definition and parameter estimation
A slecture by Chuohao Tang
Please leave me comment below if you have any questions, if you notice any errors or if you would like to discuss a topic further.
Questions and Comments
Review by S. Chatzidakis
Overall, the notes are very good and the explanation clear and easy to follow. A few comments are written below:
- At the end of section 3 "Parameter Estimation" there is a typo, it should be p(ω2)=Νω2/Ν
- In the 1D example, it is not clear how to find the separation boundary. For example, it should be mentioned that training data are used to estimate the distribution and the priors and then by equating for the two classes the boundary can be determined. Subsequently, the testing data are used to calculate the misclassification rate.
- Although the misclassification result (error rate) is provided in the 2D example, it is missing from the 1D example.
-
Back to Bayes rule in practice: definition and parameter estimation
