Machine science, the best technology

Question

I am new to machine learning. I am familiar with SVM, neural networks and GA. I would like to know how best to learn how to classify images and audio. SVM does a decent job, but takes a lot of time. Does anyone know faster and better? I would also like to know the fastest library for SVM.

Answer 1

Your question is good, and it is related to modern classification algorithms, as you say, the choice of classifier depends on your data, in the case of images I can tell you that there is one of the methods called Ada-Boost, read this and this to find out more of this, on the other hand, you can find many people who do some research, for example, in the Gender Classification of Persons using Adaboost [Rodrigo Versha, Javier Ruiz del Solne and Mauricio Correa] say:

“Adaboost-mLBP is superior to all other Adaboost-based methods, as well as basic methods (SVM, PCA and PCA + SVM).” Take a look at this.

Answer 2

- , , , VW SVM.

Answer 3

SVM

, SVN

, SVM .

stanford ML.

Answer 4

. .

Answer 5

, .

, , SIFT, , .. .

, , , MFCC.

, . , KNN .

, .

Answer 6

. , . , , , .

, , . , SVM , . , , . SVM - .

, SVD, , . , , . , SVM, .

, .