An Ensemble-based Approach to Click-Through Rate Prediction for Promoted Listings at Etsy

摘要

Etsy is a global marketplace where people across the world connect to make,buy and sell unique goods. Sellers at Etsy can promote their product listingsvia advertising campaigns similar to traditional sponsored search ads.Click-Through Rate (CTR) prediction is an integral part of online searchadvertising systems where it is utilized as an input to auctions whichdetermine the final ranking of promoted listings to a particular user for eachquery. In this paper, we provide a holistic view of Etsy's promoted listings'CTR prediction system and propose an ensemble learning approach which is basedon historical or behavioral signals for older listings as well as content-basedfeatures for new listings. We obtain representations from texts and images byutilizing state-of-the-art deep learning techniques and employ multimodallearning to combine these different signals. We compare the system tonon-trivial baselines on a large-scale real world dataset from Etsy,demonstrating the effectiveness of the model and strong correlations betweenoffline experiments and online performance. The paper is also the firsttechnical overview to this kind of product in e-commerce context.