Query Brand Entity Linking in E-Commerce Search

TL;DR

This paper tackles brand entity linking in e-commerce search queries by proposing a two-stage method and a novel end-to-end approach, addressing challenges like short queries and large brand space, validated through benchmarks and online tests.

Contribution

It introduces a new end-to-end extreme multi-class classification method for brand linking in short, unstructured e-commerce queries, improving accuracy and efficiency.

Findings

The end-to-end approach outperforms traditional methods in accuracy.

Both methods significantly improve brand retrieval in online search.

Offline benchmarks and A/B tests confirm effectiveness.

Abstract

In this work, we address the brand entity linking problem for e-commerce search queries. The entity linking task is done by either i)a two-stage process consisting of entity mention detection followed by entity disambiguation or ii) an end-to-end linking approaches that directly fetch the target entity given the input text. The task presents unique challenges: queries are extremely short (averaging 2.4 words), lack natural language structure, and must handle a massive space of unique brands. We present a two-stage approach combining named-entity recognition with matching, and a novel end-to-end solution using extreme multi-class classification. We validate our solutions by both offline benchmarks and the impact of online A/B test.