Deep Pairwise Learning To Rank For Search Autocomplete

TL;DR

This paper introduces DeepPLTR, a neural network-based pairwise ranking model that improves search autocomplete by leveraging context and behavior, achieving better offline ranking and online commercial performance.

Contribution

The paper presents a novel neural network pairwise ranker for autocomplete, outperforming LambdaMART with context-aware features and demonstrating real-world benefits.

Findings

+3.90% MRR lift offline

+0.06% GMV lift online

Effective context-aware ranking improvement

Abstract

Autocomplete (a.k.a "Query Auto-Completion", "AC") suggests full queries based on a prefix typed by customer. Autocomplete has been a core feature of commercial search engine. In this paper, we propose a novel context-aware neural network based pairwise ranker (DeepPLTR) to improve AC ranking, DeepPLTR leverages contextual and behavioral features to rank queries by minimizing a pairwise loss, based on a fully-connected neural network structure. Compared to LambdaMART ranker, DeepPLTR shows +3.90% MeanReciprocalRank (MRR) lift in offline evaluation, and yielded +0.06% (p < 0.1) Gross Merchandise Value (GMV) lift in an Amazon's online A/B experiment.