Application of Linear Discriminant Analysis and k-Nearest Neighbors Techniques to Recommendation Systems

Abstract

Among the different techniques of Machine Learning, we have selected various of them, such as SVM, CART, MLP, kNN, etc. to predict the score of a particular wine and give a recommendation to a user. In this paper, we present the results from the LDA and kNN techniques, applied to data of Rioja red wines, specifically with Rioja Qualified Denomination of Origin. Principal Component Analysis has been used previously to create a new and smaller set of data, with a smaller number of characteristics to manage, contrast, and interpret these data more easily. From the results of both classifiers, LDA and kNN, we can conclude that they can be useful in the recommendation system.