Improving E-commerce Search with Category-Aligned Retrieval

TL;DR

This paper introduces a Category-Aligned Retrieval System (CARS) that enhances e-commerce search relevance by predicting product categories from user queries and boosting within those categories, demonstrating significant improvements in category prediction accuracy.

Contribution

The paper presents a novel method for creating trainable category prototypes from query embeddings and evaluates its effectiveness with different embedding models, highlighting the importance of adaptive integration strategies.

Findings

OpenAI's embedding model increased category prediction accuracy from 43.8% to 83.2%.

Naive category boosting can negatively impact search relevance metrics.

Dataset ambiguities and system sensitivity affect the effectiveness of category-based boosting.

Abstract

Traditional e-commerce search systems often struggle with the semantic gap between user queries and product catalogs. In this paper, we propose a Category-Aligned Retrieval System (CARS) that improves search relevance by first predicting the product category from a user's query and then boosting products within that category. We introduce a novel method for creating "Trainable Category Prototypes" from query embeddings. We evaluate this method with two models: a lightweight all-MiniLM-L6-v2 and OpenAI's text-embedding-ada-002. Our offline evaluation shows this method is highly effective, with the OpenAI model increasing Top-3 category prediction accuracy from a zero-shot baseline of 43.8% to 83.2% after training. The end-to-end simulation, however, highlights the limitations of blindly applying category boosts in a complex retrieval pipeline: while accuracy is high, naive integration can negatively affect search relevance metrics such as nDCG@10. We argue that this is partly due to dataset-specific ambiguities (e.g., polysemous queries in the Amazon ESCI corpus) and partly due to the sensitivity of retrieval systems to over-constraining filters. Crucially, these results do not diminish the value of the approach; rather, they emphasize the need for confidence-aware and adaptive integration strategies.