| Line 17: | Line 17: | ||
=Questions and Comments= | =Questions and Comments= | ||
| − | * This slecture will be reviewed by Sujin Jang. | + | * (This slecture will be reviewed by Sujin Jang) In this lecture, the author introduces basic principles of NN. Then various aspects of errors and distance metrics related to NN are further described. Overall, it is well written and easy to follow the main concept. Also, proper examples are provided to aid the understanding. Especially, I like section 4 Metric Type & Application. The effect of different distance metrics in boundary shapes is well represented in Figure 3. Also, the examples on body posture and shape recognition are briefly explained but effective to get a practical sense how NN can be implemented. I recommend the author to put little bit more details on those examples or to provide more practical examples with details. Also, comparison with other classification methods will be another future direction of this lecture. |
Revision as of 07:32, 1 May 2014
Questions and comments
A slecture by Sang Ho Yoon
Partly based on the ECE662 Spring 2014 lecture material of Prof. Mireille Boutin.
If you have any questions, comments, etc. please post them on this page.
Go to Nearest Neighbor Method.
Questions and Comments
- (This slecture will be reviewed by Sujin Jang) In this lecture, the author introduces basic principles of NN. Then various aspects of errors and distance metrics related to NN are further described. Overall, it is well written and easy to follow the main concept. Also, proper examples are provided to aid the understanding. Especially, I like section 4 Metric Type & Application. The effect of different distance metrics in boundary shapes is well represented in Figure 3. Also, the examples on body posture and shape recognition are briefly explained but effective to get a practical sense how NN can be implemented. I recommend the author to put little bit more details on those examples or to provide more practical examples with details. Also, comparison with other classification methods will be another future direction of this lecture.
