Classifying multiple (2+) features

7 次查看（过去 30 天）
tafteh
tafteh 2015-10-20
回答： Walter Roberson 2015-10-20
Hi All, I have 2 (or more) feature variables in the dataset which are labeled 0 and 1 (total 2 classes). I am now using classify command in Matlab to do LDA classification. I am wondering if I use 3 or more features, would "classify" command accurately do the classification for me or not? It other words, does it classify using multiple features or only good for have 2 features in the dataset.
My other question is: how does it (classify/LDA) different from fisher's algorithm? Is it the same? Do I have to go to SVM instead all the above?
I would appreciate your assistance/guid on above. Best,

请先登录，再回答此问题。

采纳的回答

Walter Roberson
Walter Roberson 2015-10-20
classify() uses Discriminant Analysis and can handle multiple classes.
For other classification techniques, you can use the new http://www.mathworks.com/help/stats/classificationecoc-class.html which can handle any binary classifier (not just SVM)

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Statistics and Machine Learning Toolbox 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by