“…In 1984, based on principal component analysis (PCA) and ordinary least squares regression (OLS), partial leastsquares regression (PLS) was proposed by Wold et al (Wold et al, 1984), and it was one of the most effective methods for QSPR (Chen et al, 2003;Niu et al, 2005;Niu et al, 2006a;Niu et al, 2006b). Along with the development of intelligent algorithms, artificial neural network (ANN) was used to fit the nonlinearity continuous functions, and the capability of prediction was improved (Tabaraki et al, 2006;Arab Chamjangali et al, 2007).…”