Comprehensive Semantic Search for Us Case Law

Posted: 18 Jan 2023

Date Written: September 22, 2022

Abstract

Keyword search is easy to learn for our customers. It offers relatively higher precision SERP. Semantic search can leverage the SOTA deep learning technology to understand the semantic meaning of query and document to offer better results. But sometimes, semantic only search engine offers very poor SERP.

In this paper, we share our study on how to leverage both keyword and semantic search to offer optimized SERP. One US case document normally has 10k+ wordings. it's not practical to naively feed this huge volume of text to deep learning model. we also share a novel approach for handling long text semantic search.

Keywords: Semantic Serach

Suggested Citation

Zhang, Libing and Wang, Samuel and Kong, Rui and Cheng, Teng, Comprehensive Semantic Search for Us Case Law (September 22, 2022). Proceedings of the 6th Annual RELX Search Summit, Available at SSRN: https://ssrn.com/abstract=4324115

Samuel Wang

LexisNexis ( email )

P. O. Box 933
Dayton, OH 45401
United States

Rui Kong

LexisNexis ( email )

Teng Cheng

LexisNexis ( email )

P. O. Box 933
Dayton, OH 45401
United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
195
PlumX Metrics