Hierarchical gtv estimation improves container and item level transaction predictions
Hierarchical GTV Estimation: Bridging the Gap between Container-Level Ranking and Item-Level Conversion
Information Retrieval
Summary
Online shopping platforms often group items into containers like live streams or stores, but sales happen at the item level. This mismatch makes it hard to estimate the total money made from sales. The authors propose a new way to predict this total value better by using a system that looks at both container-level and item-level data, breaking down purchases into parts like how likely someone buys, how many they buy, and at what price. Their method works better than previous ones and is already used in real e-commerce platforms.
What this means in practice
- •For e-commerce platform engineers: Increase revenue predictions accuracy by integrating item-level sales data with container-level rankings in online marketplaces.
- •For digital marketing teams: Better estimate transaction outcomes of live-stream sales events to optimize marketing strategies and resource allocation.
Authors
Zerong Lan, Fan Zhang, Chuang Chen, Tianchun Huang, Teng Zhang, Xingxing Wang
Abstract
Emerging e-commerce platforms, such as live-streaming and Online-to-Offline (O2O) services, inherently operate under a "Container-based" display paradigm. In this setting, a critical granularity mismatch exists: the platform performs ranking and estimation at the container level (e.g., livestream rooms or stores), whereas the actual transactions occur at the item level within these containers. This structural discrepancy complicates Gross Transaction Value (GTV) estimation, introducing unique challenges including high label variance, severe label sparsity, and dynamic item heterogeneity. To address these challenges, we propose a novel framework named Hierarchical GTV Estimation (HGE). HGE employs a Hybrid Set-Aware Encoder (HSAE) to model the dynamic composition of items and feature interactions. To mitigate label variance, we introduce a Proxy-Label Learning (PLL) strategy that decomposes post-click GTV into purchase probability, quantity, and unit price. Furthermore, a Residual Correction and Ensemble Fusion (RCEF) module is designed to handle label sparsity by effectively combining item-level aggregation with container-level inference. Extensive experiments on large-scale industrial datasets demonstrate that HGE significantly outperforms state-of-the-art baselines, achieving a +0.63% lift in XAUC offline and a +1.42% increase in Revenue Per Search (RPS) in online A/B tests. HGE framework has been fully deployed into our main traffic.