DeText: A Deep Text Ranking Framework with BERT

TL;DR

DeText is an efficient BERT-based deep learning framework for search ranking that improves accuracy while being suitable for real-world industry applications, demonstrated through extensive offline and online experiments.

Contribution

The paper introduces DeText, a novel, open-source deep ranking framework that leverages BERT efficiently for industry-scale search systems.

Findings

Significant performance improvements over state-of-the-art methods.

Effective in both offline and online search system experiments.

Applicable to various ranking tasks in real-world scenarios.

Abstract

Ranking is the most important component in a search system. Mostsearch systems deal with large amounts of natural language data,hence an effective ranking system requires a deep understandingof text semantics. Recently, deep learning based natural languageprocessing (deep NLP) models have generated promising results onranking systems. BERT is one of the most successful models thatlearn contextual embedding, which has been applied to capturecomplex query-document relations for search ranking. However,this is generally done by exhaustively interacting each query wordwith each document word, which is inefficient for online servingin search product systems. In this paper, we investigate how tobuild an efficient BERT-based ranking model for industry use cases.The solution is further extended to a general ranking framework,DeText, that is open sourced and can be applied to various rankingproductions. Offline and online experiments of DeText on threereal-world search systems present significant improvement overstate-of-the-art approaches.