White line article recommendation system based on the KNN algorithm

Abstract

This research seeks to improve the digital marketing process for e-commerce issues, its main objective is to implement and operate a recommendation system that allows to correctly recommend a product to a customer saving time in their search and decision process. The K nearest neighbors’ algorithm and its Euclidean distance formula are used to improve the accuracy of the results. For this case we worked with the preferences of a user and a quantity of more than 100 products of different models and functionalities that are identified by identification variables such as color, brand, model, price, which are used to calculate the distance and generate "N" recommendations closer to the customer's tastes, the results show that the proposed algorithm is efficient in terms of product recommendation, generating recommendations efficiently in relation to customer preferences.