Taxonomy-based Negative Sampling In Personalized Semantic Search for E-commerce

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Jinadu, Uthman, Er, Siawpeng, Yu, Le, Liang, Chen, Li, Bingxin, Ding, Yi, Velkoski, Aleksandar
Format:	Preprint
Published:	2025
Subjects:	Information Retrieval
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

_version_	1866917054850793472
author	Jinadu, Uthman Er, Siawpeng Yu, Le Liang, Chen Li, Bingxin Ding, Yi Velkoski, Aleksandar
author_facet	Jinadu, Uthman Er, Siawpeng Yu, Le Liang, Chen Li, Bingxin Ding, Yi Velkoski, Aleksandar
contents	Large retail outlets offer products that may be domain-specific, and this requires having a model that can understand subtle differences in similar items. Sampling techniques used to train these models are most of the time, computationally expensive or logistically challenging. These models also do not factor in users' previous purchase patterns or behavior, thereby retrieving irrelevant items for them. We present a semantic retrieval model for e-commerce search that embeds queries and products into a shared vector space and leverages a novel taxonomy-based hard-negative sampling(TB-HNS) strategy to mine contextually relevant yet challenging negatives. To further tailor retrievals, we incorporate user-level personalization by modeling each customer's past purchase history and behavior. In offline experiments, our approach outperforms BM25, ANCE and leading neural baselines on Recall@K, while live A/B testing shows substantial uplifts in conversion rate, add-to-cart rate, and average order value. We also demonstrate that our taxonomy-driven negatives reduce training overhead and accelerate convergence, and we share practical lessons from deploying this system at scale.
format	Preprint
id	arxiv_https___arxiv_org_abs_2511_00694
institution	arXiv
publishDate	2025
record_format	arxiv
spellingShingle	Taxonomy-based Negative Sampling In Personalized Semantic Search for E-commerce Jinadu, Uthman Er, Siawpeng Yu, Le Liang, Chen Li, Bingxin Ding, Yi Velkoski, Aleksandar Information Retrieval Large retail outlets offer products that may be domain-specific, and this requires having a model that can understand subtle differences in similar items. Sampling techniques used to train these models are most of the time, computationally expensive or logistically challenging. These models also do not factor in users' previous purchase patterns or behavior, thereby retrieving irrelevant items for them. We present a semantic retrieval model for e-commerce search that embeds queries and products into a shared vector space and leverages a novel taxonomy-based hard-negative sampling(TB-HNS) strategy to mine contextually relevant yet challenging negatives. To further tailor retrievals, we incorporate user-level personalization by modeling each customer's past purchase history and behavior. In offline experiments, our approach outperforms BM25, ANCE and leading neural baselines on Recall@K, while live A/B testing shows substantial uplifts in conversion rate, add-to-cart rate, and average order value. We also demonstrate that our taxonomy-driven negatives reduce training overhead and accelerate convergence, and we share practical lessons from deploying this system at scale.
title	Taxonomy-based Negative Sampling In Personalized Semantic Search for E-commerce
topic	Information Retrieval
url	https://arxiv.org/abs/2511.00694