Preparation of mixture solutions and collection of commercial mango juice Standard mixture solutions of three sugars available in mango juice, i.e., glucose, fructose and sucrose, were prepared. Eight different concentrations of glucose (0.5, 1.0, 1.5, 2.0, 2.5, 3, 5, 10 percent), fructose (0.5, 1.0, 1.5, 2.0, 2.5, 3, 5, 10 percent) and sucrose (7.0, 7.5, 8.0, 8.5, 9.0, 9.5, 10.0, 15.0 percent) were used to prepare synthetic mixture solutions. Here “Orthogonal Experimental Design” was used to statistically maximize the information in the outputs. Thus, in total 64 mixtures were prepared with different concentrations of glucose, fructose and sucrose. According to the combination of concentrations from experimental design, the sugars were dissolved into de-ionized water to make solutions. Next, we collected 15 commercially available mango juices of different locally manufacturing companies. Then concentrations of glucose, fructose and sucrose in commercial mango juices were measured at laboratory by standard AOAC method (Horwitz, 2005). Both mixture solutions and commercial juices were used in scientific instrument, Fourier Infrared (FTIR) spectrophotometer, to get spectral data from the instrument. Finally, known concentrations of simple sugars and spectral data of synthetic mixture solutions were used to develop and validate a method, and spectra of real mango juices were used to test the method for prediction of glucose, fructose and sucrose in mango juices and classifying them. FTIR measurements In Fourier Transform Infrared (FTIR) spectroscopy, IR radiation is passed through a sample. Some of the infrared radiation is absorbed by the sample and some of it is passed through (transmitted). The resulting spectrum represents the molecular absorption and transmission, creating a molecular fingerprint of the sample. Like a fingerprint no two unique molecular structures produce the same infrared spectrum. This makes infrared spectroscopy useful for several types of analysis. FTIR spectrometer (Shimadzu, Model: IRAfinity1) connected to software of IRSolution Operating system (Version 1.40) was used to obtain FTIR spectra of samples. The samples were placed in contact with Attenuated Total Reflectance (ATR) element at controlled ambient temperature. Finally, the mixture solutions and real mango juices were run in FTIR to get their respective spectra. FTIR spectra were collected in frequency 4000-650 cm-1 by co-adding 30 scans and at resolution of 4 cm-1. All spectra were rationed against a background of air spectrum. Before every scan, a new reference air background spectrum was taken. There spectra were recorded as absorbance values at each data point in triplicate. The ATR plate was carefully cleaned in situ by wiping it with acetone, and dried with soft tissue before filling in with next sample. Preprocessing of spectral data The spectral data acquired from instrument contain spectra background information and noises which are interfered desired relevant quality attributes information. Interfering spectral parameters, such as light scattering, path length variations and random noise resulted from variable physical sample properties or instrumental effects need to be eliminated or reduced in order to obtain reliable, accurate and stable calibration models. Thus, it is very necessary to pre-process spectral data prior to modeling (Rinnan et al., 2009). From spectra of FTIR we can see that there is no spectral peak and almost no variance of absorbance below the wave number 3700 cm-1 contain least information for prediction. So, wave number range 3700-648 cm-1 has been selected for further analysis. Here, spectral data were de-noised with Savitzky–Golay filtering (Palma et al., 2002; Nicolai et al., 2006) is used to de-noised the spectral data. Classes of mango juice Concentrations of glucose, fructose and sucrose in commercial mango juices were measured at laboratory by standard AOAC method. Average concentration of glucose, fructose and sucrose are 1.6%, 1.7% and 8.4% respectively. So, for the sake of classification we divide the simple sugar concentrations into two groups (Group 1 and Group 2) on the basis of approximation of average concentration of the simple sugars in commercial mango juice. Chemometric techniques for classification It is very often necessary to classify foods on the basis of their raw materials, species of ingredients, brand, geographical origin, category of product etc. to identify their specialty or traceability. This is one the process to identify the authenticity of food. Several classification techniques are used in food authentication studies by food scientists and food quality controlling authorities. Here, ANN and PLS-DA have been used to develop classification model to classify commercial mango juice as adulterated or safe on the basis of simple sugars they contain. Here the inputs are spectral data of mango juices and classification categories (high and low level of concentrations of sugars) are outputs. Artificial Neural Network (ANN) An artificial neural network (ANN) is a data processing system based on the structure of the biological neural simulation by learning from the data generated experimentally or using validated models (Bhotmange and Shastri, 2011). A network consists of a sequence of layers with connections between successive layers. Data to the network is presented at input layer and the response of the network to the given data is produced in the output layer. There may be several layers between these two principal layers, which are called hidden layers. Finally, neural network approach build a predictive model for quantification or classification in a complex system.