Background: There are limitations in determining the content and functional compositions of algae in different species, season and coastal environments. Therefore, the demand for screening of compositions of species is very important for easy access, finding properties and information in algae. FTIR can reveal species-specific spectral signatures and spectral features, and UV-Vis can introduce information about their pigments and metabolites. By combination of spectroscopic data with chemometrics techniques it is possible to accurate estimate about different functional groups and composition of algae without quantification of cell contents. Based on the findings, diversity of algae species without time consuming sample preparation, measuring cell content and large consumption of material for analysis was done. Aim: The main goal of present study is to illustrate advantageous of combination of spectroscopic data with chemometrics approaches for discrimination and identification of marine algae species which harvested from different coastal environments. Also, the dimensionality reduction by clustering of variables obtained by Kohonen self-organization map (SOMs) is our interest. Combining different spectroscopic data create synergistic information about different functional groups and pigment of bioactive phytochemicals and hence, enhanced the classification measures. Finally, the fusion matrix was built using the SOMs clusters of FTIR and UV-Vis data.
Methodology: The data sets were pretreated with autoscaling and Savitzky-Golay smoothing for elimination of the noise and background interferences and improving the accuracy of classification. Also, the Kohonen self-organization map (SOMs) was considered for reducing the dimension of data. The principal component analysis (PCA), PCA- discriminant analysis (PCA-DA), and PLS (partial least squares)-DA were employed for discrimination. The synergistic advantages were considered by data fusion strategy. The fusion matrix w
