Idea
A generative AI framework integrating search and recommendation to improve user engagement and conversion for digital platforms.
Research Paper
Core Innovation
This paper introduces IntSR, a novel generative framework that integrates search and recommendation tasks by leveraging distinct query modalities. It uniquely addresses the challenges of computational complexity and dynamic corpus changes in integrated S&R systems. Unlike prior work focusing only on retrieval and ranking, IntSR treats queries as central elements to unify disparate S&R behaviors.
Market Size (TAM)
$20–50B TAM for search and recommendation platforms; $2–10B SAM from e-commerce, travel, and location-based services. Driven by growing digital content and user engagement demands.
Potential Customers & Pain Points
- E-commerce Platforms Needing Unified Search and Recommendation
- Digital Asset Managers Seeking Higher GMV
- Location-Based Service Providers Improving POI Recommendations
- Travel Apps Enhancing Mode Suggestions Accuracy
Business Model
SaaS platform licensing to digital service providers with tiered pricing based on usage and integration complexity.
Competitive Landscape
- Google Recommendations AI
- Amazon Personalize
- Microsoft Azure Cognitive Search
Implementation Challenges
- High computational resource requirements
- Complex integration of diverse query modalities
- Dynamic data corpus management challenges
Validation Strategy
- Pilot deployment with select e-commerce and travel partners
- Measure key metrics like GMV
- CTR
- and accuracy improvements
- Iterate model based on real-world feedback and scalability tests
Research Paper Overview
IntSR: An Integrated Generative Framework for Search and Recommendation
Summary
Generative recommendation has emerged as a promising paradigm, demonstrating remarkable results in both academic benchmarks and industrial applications. However, existing systems predominantly focus on unifying retrieval and ranking while neglecting the integration of search and recommendation (S&R) tasks. What makes search and recommendation different is how queries are formed: search uses explicit user requests, while recommendation relies on implicit user interests. As for retrieval versus ranking, the distinction comes down to whether the queries are the target items themselves. Recognizing the query as central element, we propose IntSR, an integrated generative framework for S&R. IntSR integrates these disparate tasks using distinct query modalities. It also addresses the increased computational complexity associated with integrated S&R behaviors and the erroneous pattern learning introduced by a dynamically changing corpus. IntSR has been successfully deployed across various scenarios in Amap, leading to substantial improvements in digital asset's GMV(+3.02%), POI recommendation's CTR(+2.76%), and travel mode suggestion's ACC(+5.13%).