Automatic Non-Structural Protien 1 recognition based on LDA classifier

This paper discusses the possibilities of Non Structural Protein1 (NS1) fingerprint can be classified from Raman spectra of saliva using Linear Discriminant Analysis (LDA). LDA is a supervised statistical method that can be used to classify two or more groups of data. In this research, Raman spectra of saliva and NS1-saliva mixture are obtained using Surface Enhanced Raman Spectroscopy (SERS) technique where gold coated slides (GS) are used as substrates. The NS1-saliva mixtures are prepared into different concentration of 10ppm, 50ppm and 100ppm. After applying simple LDA algorithm, the transformed data of saliva and NS1-saliva mixture are overlapping. However the overlapped data is reduced as the concentration of the mixture increase. It indicate that the algorithm is more suitable for samples with higher amount of NS1. Integration of LDA with other algorithm need to be considered for better classification.