BroadGen: A Framework for Generating Effective and Efficient Advertiser Broad Match Keyphrase Recommendations

2025-05-25

AI Summary

Sponsored search keyword recommendation under broad match faces challenges including imprecise targeting, scarce supervision signals, and high operational overhead. Method: This paper formally defines the “ideal broad match” via a dual-criterion standard—high relevance and temporal stability—and proposes BroadGen, a framework that jointly performs large-scale keyword generation and filtering through token correspondence modeling and historical query mining. Contribution/Results: Deployed in eBay’s production environment, BroadGen supports daily recommendations for millions of sellers and over 2.3 billion items, significantly improving matching accuracy and query coverage while reducing advertisers’ manual maintenance efforts. Its core innovations lie in (i) establishing a theoretically grounded evaluation criterion for broad match and (ii) introducing an end-to-end generative paradigm that balances efficiency and effectiveness.

Technology Category

• Search and Optimization: Algorithm Configuration
• Natural Language Processing: Generation
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Search and Retrieval-Augmented AI: Ad search and search for Web retail
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions

Abstract

In the domain of sponsored search advertising, the focus of Keyphrase recommendation has largely been on exact match types, which pose issues such as high management expenses, limited targeting scope, and evolving search query patterns. Alternatives like Broad match types can alleviate certain drawbacks of exact matches but present challenges like poor targeting accuracy and minimal supervisory signals owing to limited advertiser usage. This research defines the criteria for an ideal broad match, emphasizing on both efficiency and effectiveness, ensuring that a significant portion of matched queries are relevant. We propose BroadGen, an innovative framework that recommends efficient and effective broad match keyphrases by utilizing historical search query data. Additionally, we demonstrate that BroadGen, through token correspondence modeling, maintains better query stability over time. BroadGen's capabilities allow it to serve daily, millions of sellers at eBay with over 2.3 billion items.

Problem

• Addresses limitations of exact match keyphrase recommendations in ads
• Improves targeting accuracy for broad match keyphrase recommendations
• Ensures query relevance and stability in dynamic search environments

Innovation

• Utilizes historical search query data
• Employs token correspondence modeling
• Ensures query relevance and stability