Zu dieser Karteikarte gibt es einen kompletten Satz an Karteikarten. Kostenlos!
69
How does the number of neighbors influence k-NN?
If you only look at a small number of neighbors the decision boundary between classes is very distinct.
A large value of k reduces the effect of noise on the classification but the boundary between classses becomes less distinct.
The best choice of k depends on the data.
A large value of k reduces the effect of noise on the classification but the boundary between classses becomes less distinct.
The best choice of k depends on the data.
Tags:
Quelle: CI Teil 1 Lecture 6
Quelle: CI Teil 1 Lecture 6
Karteninfo:
Autor: Sepp Samuel
Oberthema: Telematik
Thema: Computational Intelligence
Schule / Uni: TU Graz
Veröffentlicht: 02.07.2014