Abstract
Digital advertising may be challenged by latency in manual workflows and a semantic gap between fast-moving cultural trends and structured product data, potentially resulting in missed commercial opportunities. A described system can address these challenges using an automated, multi-agent orchestration workflow. For example, a surveillance agent can identify breakout trends from various data streams, such as social media feeds or search queries. These trends may then be semantically expanded using a large language model. The expanded description can be converted into a vector embedding to perform a similarity search against a vector database of a product catalog, which may identify relevant products. After automated governance and policy checks, the system can initiate a responsive advertising campaign. This process may reduce the delay between trend detection and commercial action and can improve the relevance of product matching for trend-based consumer interest.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Arora, Jasbeer; Tripathi, Anurag; and Singh, Gurvinder, "Autonomous Orchestration of Trend-Driven Advertising Using Semantic Product Matching", Technical Disclosure Commons, (August 26, 2026)
https://www.tdcommons.org/dpubs_series/11484