English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Learning to Rank for Maps at Airbnb (KDD 2024)

Forum topic · 小凯 · 2026-07-05

Summary

This KDD 2024 paper from Airbnb presents a learning-to-rank (LTR) approach for map-based search. Airbnb's map experience lets users browse listings spatially, which introduces ranking challenges distinct from traditional list-based search: results must be relevant both as individual items and as a coherent set displayed within the visible map viewport. The work addresses how to apply learning-to-rank techniques to this map-centric ranking scenario, covering the system design, model training, and evaluation within Airbnb's production search infrastructure. As with other industrial LTR systems, the approach draws on training signals derived from user interactions such as bookings and clicks, and must balance ranking quality with latency, scalability, and product constraints inherent to a live marketplace. The paper offers practitioners a case study of deploying machine-learned ranking for a geospatial browsing surface at scale, and situates itself within the broader verticals and search/recommendation literature. Readers interested in applied ranking systems, geospatial search, and industrial machine learning at Airbnb will find the full details, including quantitative results, in the official ACM publication (DOI: 10.1145/3637528.3671648). Note: quantitative results should be verified against the original PDF.

Learning to Rank for Maps at Airbnb (KDD 2024)

Source: https://dl.acm.org/doi/abs/10.1145/3637528.3671648 Venue: KDD 2024 | Category: Verticals / Search & Ranking

Overview

This paper describes Airbnb's work on applying Learning to Rank (LTR) to its map product. Unlike conventional list-based search, Airbnb's map experience requires ranking listings that are browsed spatially: the ranking model must consider not only the relevance of individual listings but also how results are presented within the visible map region a user is exploring.

Key points

Glossary

| Term | Meaning | |------|---------| | LTR | Learning to Rank | | nDCG | Normalized Discounted Cumulative Gain, a ranking quality metric | | Ranking cascade | Pipeline of candidate retrieval followed by fine-grained re-ranking |

Recommended reading paths

1. Researchers: Compare the map-ranking formulation with standard LTR benchmarks and check how statistical significance and compute costs are reported. 2. Engineers: Identify reusable components (encoders, re-rankers) and evaluate integration cost with existing search stacks. 3. Product managers: Focus on user-perceivable benefits (latency, result relevance on the map) rather than offline metrics alone.

Tags

#learning-to-rank#airbnb#kdd-2024#search-ranking#maps#information-retrieval#industrial-ml#recommender-systems

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178209025