Showing 1–1 of 1 results for author: Shekasta, M

New Item Consumption Prediction Using Deep Learning

Authors: Michael Shekasta, Gilad Katz, Asnat Greenstein-Messica, Lior Rokach, Bracha Shapira

Abstract: Recommendation systems have become ubiquitous in today's online world and are an integral part of practically every e-commerce platform. While traditional recommender systems use customer history, this approach is not feasible in 'cold start' scenarios. Such scenarios include the need to produce recommendations for new or unregistered users and the introduction of new items. In this study, we pres… ▽ More Recommendation systems have become ubiquitous in today's online world and are an integral part of practically every e-commerce platform. While traditional recommender systems use customer history, this approach is not feasible in 'cold start' scenarios. Such scenarios include the need to produce recommendations for new or unregistered users and the introduction of new items. In this study, we present the Purchase Intent Session-bAsed (PISA) algorithm, a content-based algorithm for predicting the purchase intent for cold start session-based scenarios. Our approach employs deep learning techniques both for modeling the content and purchase intent prediction. Our experiments show that PISA outperforms a well-known deep learning baseline when new items are introduced. In addition, while content-based approaches often fail to perform well in highly imbalanced datasets, our approach successfully handles such cases. Finally, our experiments show that combining PISA with the baseline in non-cold start scenarios further improves performance. △ Less

Submitted 12 May, 2019; v1 submitted 5 May, 2019; originally announced May 2019.