# Talk:Nearest feature line

Jump to: navigation, search

1. In the first paragraph, the authors mentioned that the NFL method effectively improves the classification performance especially when the number of templates per class is small (do you mean small sample size?). However, they did not discuss this in the text.

2. “Motivation” section: change “The nearest neighbor (NN)classification…” to “The nearest neighbor (NN) classification…”.

3. “Overview of NFL” section: the second paragraph, last sentence, according to the definition of the distance between a point and a line, suggest to change the sentence “…the distance between the query and its projection onto the FL generated between FLs (should be FPs?)” to “…the distance between the query and the FLs”

4. “Previous work” section, the third paragraph, second line, typo: {\em linear object classes}

5. “Results” section, “Also, the NFL outperforms k-NN…”, here k=?