Abstract

Entity Resolution (ER) is defined as the algorithmic problem of determining when two or more entities refer to the same underlying entity. In the e-commerce domain, the problem tends to arise when the same product is advertised on multiple platforms, but with slightly (or even very) different descriptions, prices and other attributes. While ER has been well-explored for domains like bibliographic citations, biomedicine, patient records and even restaurants, work on product ER is not as prominent. In this paper, we report preliminary results on an unsupervised product ER system that is simple and extremely lightweight. The system is able to reduce mean rank reductions on some challenging product ER benchmarks by 50-70% compared to a text-only benchmark by leveraging a combination of text and neural graph embeddings.

Date

March 15, 2026

Authors

Mozhdeh Gheini, Mayank Kejriwal

Conference

eCOM@ SIGIR