CPL - Chalmers Publication Library
| Utbildning | Forskning | Styrkeområden | Om Chalmers | In English In English Ej inloggad.

Feature Reduction Based on Sum-of-SNR (SOSNR) Optimization

Yinan Yu (Institutionen för signaler och system, Signalbehandling) ; Tomas McKelvey (Institutionen för signaler och system, Signalbehandling) ; S.Y. Kung
The 39th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP) (1520-6149). p. 6756-6760. (2014)
[Konferensbidrag, refereegranskat]

Dimensionality reduction plays an important role in machine learning techniques. In classification, data transformation aims to reduce the number of feature dimensions, whereas attempts to enhance the class separability. To this end, we propose a new classifier-independent criterion called 'Sum-of-Signal-to-Noise-Ratio' (SoSNR). A framework designed for maximization with respect to this criterion is presented and three types of algorithms, respectively based on (1) gradient, (2) deflation and (3) sparsity, are proposed. The techniques are conducted on standard UCI databases and compared to other related methods. Results show trade-offs between computational complexity and classification accuracy among different approaches.

Nyckelord: classification; feature reduction; Fisher's Score; SODA; Sum-of-SNR

Article number 6854908

Den här publikationen ingår i följande styrkeområden:

Läs mer om Chalmers styrkeområden  

Denna post skapades 2014-07-07. Senast ändrad 2015-01-05.
CPL Pubid: 200239


Läs direkt!

Lokal fulltext (fritt tillgänglig)

Länk till annan sajt (kan kräva inloggning)