TY - JFULL AU - Nahid Ghasemi and Mohammad Goodarzi and Morteza Khosravi PY - 2009/9/ TI - Principal Component Analysis-Ranking as a Variable Selection Method for the Simultaneous Spectrophotometric Determination of Phenol, Resorcinol and Catechol in Real Samples T2 - International Journal of Computer and Information Engineering SP - 2133 EP - 2139 VL - 3 SN - 1307-6892 UR - https://publications.waset.org/pdf/10991 PU - World Academy of Science, Engineering and Technology NX - Open Science Index 32, 2009 N2 - Simultaneous determination of multicomponents of phenol, resorcinol and catechol with a chemometric technique a PCranking artificial neural network (PCranking-ANN) algorithm is reported in this study. Based on the data correlation coefficient method, 3 representative PCs are selected from the scores of original UV spectral data (35 PCs) as the original input patterns for ANN to build a neural network model. The results obtained by iterating 8000 .The RMSEP for phenol, resorcinol and catechol with PCranking- ANN were 0.6680, 0.0766 and 0.1033, respectively. Calibration matrices were 0.50-21.0, 0.50-15.1 and 0.50-20.0 μg ml-1 for phenol, resorcinol and catechol, respectively. The proposed method was successfully applied for the determination of phenol, resorcinol and catechol in synthetic and water samples. ER -