Principal Component Analysis PCA

I need to use PCA for features selection. Foe example if the number of features = 300 the result by PCA = 15. I used PCA command in matlab but the result complete transformation of the feature (300), but I need only the important features to save the accuracy.

[Image Analyst]

OK good luck. Do you have a question (rather than an announcement)? If so, you forgot to attach your code and code. So do that then please ask your question after reading this: http://www.mathworks.com/matlabcentral/answers/6200-tutorial-how-to-ask-a-question-on-answers-and-get-a-fast-answer

Answer 1

[KSSV]

pca gives you out score of each feature. You sort the scores and pick the first fifteen scores and the respective features.

Answer 2

[Sulaymon Eshkabilov]

Here are two nice documentation that address your raised questions:

https://www.mathworks.com/help/stats/pca.html

https://www.mathworks.com/help/stats/partial-least-squares-regression-and-principal-components-regression.html