Beyond Lexical: A Semantic Retrieval Framework for Textual SearchEngine
Information Retrieval
2020-08-11 v1
Abstract
Search engine has become a fundamental component in various web and mobile applications. Retrieving relevant documents from the massive datasets is challenging for a search engine system, especially when faced with verbose or tail queries. In this paper, we explore a vector space search framework for document retrieval. Specifically, we trained a deep semantic matching model so that each query and document can be encoded as a low dimensional embedding. Our model was trained based on BERT architecture. We deployed a fast k-nearest-neighbor index service for online serving. Both offline and online metrics demonstrate that our method improved retrieval performance and search quality considerably, particularly for tail
Cite
@article{arxiv.2008.03917,
title = {Beyond Lexical: A Semantic Retrieval Framework for Textual SearchEngine},
author = {Kuan Fang and Long Zhao and Zhan Shen and RuiXing Wang and RiKang Zhour and LiWen Fan},
journal= {arXiv preprint arXiv:2008.03917},
year = {2020}
}
Comments
9 pages