Pre-training Tasks for User Intent Detection and Embedding Retrieval in E-commerce Search

TL;DR

This paper introduces specialized pre-training tasks tailored for user intent detection and embedding retrieval in e-commerce search, resulting in smaller, more effective models that outperform baselines in industrial settings.

Contribution

The authors develop novel pre-training tasks for specific e-commerce search modules, creating compact models that enhance performance over general BERT-based approaches.

Findings

Customized pre-trained models are less than 10% the size of BERT-base.

Proposed models significantly outperform no pre-training and general BERT fine-tuning baselines.

Open-sourced datasets facilitate reproducibility and future research.

Abstract

BERT-style models pre-trained on the general corpus (e.g., Wikipedia) and fine-tuned on specific task corpus, have recently emerged as breakthrough techniques in many NLP tasks: question answering, text classification, sequence labeling and so on. However, this technique may not always work, especially for two scenarios: a corpus that contains very different text from the general corpus Wikipedia, or a task that learns embedding spacial distribution for a specific purpose (e.g., approximate nearest neighbor search). In this paper, to tackle the above two scenarios that we have encountered in an industrial e-commerce search system, we propose customized and novel pre-training tasks for two critical modules: user intent detection and semantic embedding retrieval. The customized pre-trained models after fine-tuning, being less than 10% of BERT-base's size in order to be feasible for cost-efficient CPU serving, significantly improve the other baseline models: 1) no pre-training model and 2) fine-tuned model from the official pre-trained BERT using general corpus, on both offline datasets and online system. We have open sourced our datasets for the sake of reproducibility and future works.