arXiv 2512.02436

Semantic Trading: Agentic AI for Clustering and Relationship Discovery in Prediction Markets

By Agostino Capponi, Alfio Gliozzo, et al.

Published 2025-12-02

Citation lineage

Review the prior work and downstream research connected to this paper.

Prediction markets allow users to trade on outcomes of real-world events, but are prone to fragmentation through overlapping questions, implicit equivalences, and hidden contradictions across markets. We present an agentic AI pipeline that autonomously (i) clusters markets into coherent topical groups using natural-language understanding over contract text and metadata, and (ii) identifies within-cluster market pair…

View the original paper on arXiv