Data mining and optimisation issues in the food industry

Abstract

Data mining applications in the food industry has until now expressed in many ways, on both technical and economic terms. The most important methodologies being used are clustering, classification, feature selection and outlier detection. The techniques commonly used in data mining are artificial neural networks, decision trees, k-means type algorithms, genetic algorithms, nearest neighbour method, and rule induction. Successful case studies of the implementation of these methodologies are fruit and vegetable classification, with special focus on apples, citrus, strawberries, table olives, onions etc. the same rationale is being followed for the evaluation of processed foodstuff. Efficient solutions have been provided for wine classification, based on organoleptic characteristics, fish and meat classification, as well as robotic harvesting. Finally, applications on supply chain management and e-commerce have provided significant solutions on issues of monitoring and evaluation of dynamically changing datasets.