Stageboost: Recommending Signals Based on Counterfactual Estimation
Information RetrievalArtificial Intelligence
Summary
The authors study small text or pictures called signals that appear on eBay's item pages to help buyers decide what to purchase. They built a two-step computer model using xgboost, a type of machine learning, to choose which signals to show. Their method slightly improved the total sales and had a bigger boost for parts and accessories, mainly by helping sell more expensive items. This suggests their model helps customers make smarter purchase choices.
Authors
Darpan Singhal, Matan Mandelbrod, Tal Franji, Manasa Kolla, Vipul Gaba, Yuri Brovman
Abstract
Signals are short textual or visual snippets displayed on the eBay View-Item (VI) page, providing additional, contextual information for users about the viewed item. The aim of displaying these signals is to facilitate intelligent purchase and to incentivize engagement. In this paper, we present a 2 stage xgboost based model that optimally populates the VI page with signals. This approach has shown a 0.08% lift in overall GMB (Gross Merchandise Bought) and 0.58% increase in Parts and Accessories GMB, primarily due to increase in conversion of high average price items in online experimentation.