Classification depend on linear discriminant analysis using desired outputs
- 1 December 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 328-332
Abstract
Linear Discriminant Analysis (LDA) is a popular technique in pattern recognition. This paper presents a linear discriminant analysis for classification. Firstly, linear discriminant Analysis is introduced. The optimum design procedure is demonstrated in order to calculate the error minimization. Next, linear classifier based on desired outputs is derived to get the desired classification. Also, the performance measure is described and calculated to show the effectiveness of the classification algorithm. The usefulness of the proposed procedures was proved in the simulation results.Keywords
This publication has 5 references indexed in Scilit:
- Quantum discriminant analysis for dimensionality reduction and classificationNew Journal of Physics, 2016
- Feature selection for classification of BGP anomalies using Bayesian models2012
- Locally linear discriminant analysis for multimodally distributed classes for face recognition with a single model imageIEEE Transactions on Pattern Analysis and Machine Intelligence, 2005
- A new LDA-based face recognition system which can solve the small sample size problemPattern Recognition, 2000
- Recursive dimensionality reduction using Fisher’s linear discriminantPattern Recognition, 1998