HierCat: Hierarchical Query Categorization from Weakly Supervised Data at Facebook Marketplace

TL;DR

HierCat is a hierarchical query categorization system for Facebook Marketplace that leverages weakly supervised data and multi-task pre-training to improve search relevance and user engagement.

Contribution

The paper introduces HierCat, a novel hierarchical query categorization approach that effectively learns from weakly supervised data using multi-task pre-training and hierarchical inference.

Findings

Outperforms popular methods in offline experiments

Achieves 1.4% improvement in NDCG

Increases searcher engagement by 4.3% in online A/B tests

Abstract

Query categorization at customer-to-customer e-commerce platforms like Facebook Marketplace is challenging due to the vagueness of search intent, noise in real-world data, and imbalanced training data across languages. Its deployment also needs to consider challenges in scalability and downstream integration in order to translate modeling advances into better search result relevance. In this paper we present HierCat, the query categorization system at Facebook Marketplace. HierCat addresses these challenges by leveraging multi-task pre-training of dual-encoder architectures with a hierarchical inference step to effectively learn from weakly supervised training data mined from searcher engagement. We show that HierCat not only outperforms popular methods in offline experiments, but also leads to 1.4% improvement in NDCG and 4.3% increase in searcher engagement at Facebook Marketplace Search in online A/B testing.