# A unified Neural Network Approach to E-CommerceRelevance Learning

arXiv: https://arxiv.org/abs/2104.12302 · HTML: https://mlsanity.com/paper/2104.12302/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Apr 2021.

## Built on by (most cited first, 7 of 7)

- [Large Language Models for Relevance Judgment in Product Search](https://mlsanity.com/paper/2406.00247/) (2024)
- [Deep Bag-of-Words Model: An Efficient and Interpretable Relevance Architecture for Chinese E-Commerce](https://mlsanity.com/paper/2407.09395/) (2024)
- [Centrality-aware Product Retrieval and Ranking](https://mlsanity.com/paper/2410.15930/) (2024)
- [SPM: Structured Pretraining and Matching Architectures for Relevance Modeling in Meituan Search](https://mlsanity.com/paper/2308.07711/) (2023)
- [Improving Text Matching in E-Commerce Search with A Rationalizable, Intervenable and Fast Entity-Based Relevance Model](https://mlsanity.com/paper/2307.00370/) (2023)
- [Improving Topic Relevance Model by Mix-structured Summarization and LLM-based Data Augmentation](https://mlsanity.com/paper/2404.02616/) (2024)
- [NEAR2: A Nested Embedding Approach to Efficient Product Retrieval and Ranking](https://mlsanity.com/paper/2506.19743/) (2025)

## Abstract (arXiv)

Result relevance scoring is critical to e-commerce search user experience. Traditional information retrieval methods focus on keyword matching and hand-crafted or counting-based numeric features, with limited understanding of item semantic relevance. We describe a highly-scalable feed-forward neural model to provide relevance score for (query, item) pairs, using only user query and item title as features, and both user click feedback as well as limited human ratings as labels. Several general enhancements were applied to further optimize eval/test metrics, including Siamese pairwise architecture, random batch negative co-training, and point-wise fine-tuning. We found significant improvement over GBDT baseline as well as several off-the-shelf deep-learning baselines on an independently constructed ratings dataset. The GBDT model relies on 10 times more features. We also present metrics for select subset combinations of techniques mentioned above.