Principle Component Analysis

Friends,
I understood explanation given on PCA in help. But i am not able to understand how to select the best features/variables. Can anybody help me in this regard ?
If you have an example with data please send it to bbhatpramod@gmail.com. Your help would be greatly appreciated.
Thanq..

回答 (1 件)

Walter Roberson
Walter Roberson 2012 年 2 月 5 日

0 投票

The "best" features or variables are the ones that have the most "explanatory power" -- the ones that come out highest in PCA.

1 件のコメント

Wu Ying
Wu Ying 2015 年 2 月 12 日
for the variables, it may be a positive or negative contribute to the principle component. How to choose the highest? Just the absolute value or ?

サインインしてコメントする。

カテゴリ

ヘルプ センター および File ExchangeDimensionality Reduction and Feature Extraction についてさらに検索

質問済み:

2012 年 2 月 5 日

コメント済み:

2015 年 2 月 12 日

Community Treasure Hunt

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

Start Hunting!

Translated by