Efficient Conversion Prediction in E-Commerce Applications with Unsupervised Learning
Péter Gábor Szabó, Béla Genge · 2020
Unsupervised machine learning became a ubiquitous method appearing in E-commerce solutions that strive to provide personalized recommendations for their users. Most of those solutions embrace collaborative filtering (CF) to predict conversions, which are the beneficial user events, such as a purchase. Traditionally, the predictions were made based on rating data. However, e-commerce users seldom leave ratings. Instead, we must rely on user events, such as viewing an item or adding it to the cart. The event-based approach seems counter-intuitive, for the reason that the operation time of recommender systems increases exponentially with the increase of data-points.One of the main contributions of this paper is the UX value function. It reduces all events between an item and a user to a single user experience number, which also depends on the sequentiality of the events. We present a method to calculate this number in linear time. Then we use a deep neural network to predict the likelihood of conversions based on this number to prove the practical solvability of the problem in a scalable manner, with a relatively fast learning speed and good prediction accuracy. We have conducted an extensive experimental analysis on Kechinov’s ‘eCommerce Events History in Cosmetics Shop’ dataset, containing 8,738,120 user events. The results of those experiments prove the efficiency and applicability of the developed approach.