An empirical analysis of deep learning architecture towards Classification and prediction of food Adulteration

Due to food related disease in growing tremendously, it has become important to identify the food contamination using wide variety of analysis. The process employed for detection of food adulteration is critical task. Due to its importance, many data mining technique has been employed to identify the food adulteration based on various factors. Even there exist some challenges because most of method exhibit static behaviour. The ultimate goal of this study is to carry out of the empirical analysis of food adulteration using deep learning techniques. The deep learning technique is applied to the information leveraged from data sources. The model builds the class from the analysis of data through the information extracted from the foods substance on various properties along the ingredients and adulterants. The deep learning technique is used for representation of the classification model in order to generate feature representation of mandatory ingredients of food and to predict the contamination on the samples which produces the variance from the ingredient features in the food to identify impurities. The classification model is capable of categorizing the adulteration into multiple categories on deviation of the feature value of the each instance. Finally it is used as method for detection of the food adulteration on wide varieties. It can be used as food adulteration assessment tool to understand the risk of the food prepared.