Pre-training Tasks for User Intent Detection and Embedding Retrieval in E-commerce Search
Authors:
Open Access
- 17 October 2022
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 4424-4428
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 tech- nique 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 detec- tion 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 1 for the sake of reproducibility and future works.Keywords
All Related Versions
This publication has 17 references indexed in Scilit:
- A Joint Model of Clinical Domain Classification and Slot Filling Based on RCNN and BiGRU-CRF2019
- Billion-Scale Similarity Search with GPUsIEEE Transactions on Big Data, 2019
- Small and Practical BERT Models for Sequence Labeling2019
- Multi-passage BERT: A Globally Normalized BERT Model for Open-domain Question Answering2019
- GCDT: A Global Context Enhanced Deep Transition Architecture for Sequence Labeling2019
- E-commerce Product Query Classification Using Implicit User’s Feedback from Clicks2018
- Deep Learning for Extreme Multi-label Text Classification2017
- Amazon Search2016
- Deep Metric Learning Using Triplet Network2015
- Learning deep structured semantic models for web search using clickthrough data2013