Análisis de clusterización basado en el algoritmo K- Means Neighbourhood para la clasificación de tipos de clientes web en una empresa de retail e-commerce.

Abstract

In the modern world and with digitalization as a key pillar of our daily lives, knowingand implementing the new technologies that are being developed is of vital importancebecause they are able to streamline or improve the processes we carry out in our dailyactivities. Undoubtedly the heyday of artificial intelligence and its derivatives is the trend that is developing today, and in the business, field is no exception being infiniteits applications. Knowing the preferences of your customers without having to have them present represents a great competitive advantage over the competition, years agothis task would be seen as something impossible, but today with the implementation of new technologies it has gone from being a dream to a reality. In this work we intendto demonstrate how the implementation of a Machine Learning algorithm system would allow Retail companies that are present in the E-commerce to facilitate the classification of their customers based on similar characteristics, this through the K- means Neighbor model and for validation through the decision tree model, This will result in groupings of similar clusters based on variables such as the type of payment,the type of shipment, the use or not of coupons, the value of the shipment and the number of transactions and thus be able to subsequently provide discounts, promotionsor campaigns targeted in a better way to customers that could be more attractive to them.