Prediction Markets Based on Proximity
Move beyond binary outcomes. Our protocol uses vector embeddings to create prediction markets where accuracy is measured by proximity to reality across semantic, structural, and Euclidean dimensions.
Three Proximity Frameworks
Our protocol supports multiple distance metrics, allowing markets to capture nuanced predictions beyond simple yes/no outcomes.
Semantic Proximity
Measures similarity in meaning using natural language embeddings. Predictions are scored based on conceptual alignment with actual outcomes.
Structural Proximity
Evaluates graph-based relationships and network structures. Captures how entities relate to each other in complex systems.
Euclidean Proximity
Traditional distance metrics in vector space. Direct mathematical measurement of prediction accuracy across dimensions.
Vector Embeddings Drive Accuracy
Traditional prediction markets force complex phenomena into binary outcomes. This creates information loss and reduces market efficiency.
Our protocol uses high-dimensional vector embeddings to represent both predictions and outcomes. When reality unfolds, we measure the distance between predicted and actual vectors across multiple dimensions.
This approach enables markets for inherently non-binary questions: "How will the climate evolve?", "What will the economic landscape look like?", or "Where will the next disaster occur?"
[0.82, 0.45, 0.91, ...][0.79, 0.52, 0.88, ...]Lower distance = higher accuracy = greater payout
Applications Across Domains
Spatial prediction markets unlock forecasting capabilities for complex, multi-dimensional phenomena.
Disaster Response
Predict disaster impact zones and resource needs with geographic precision.
Market Forecasting
Forecast economic indicators across multiple correlated dimensions.
Health Outcomes
Model complex health scenarios with multi-factor outcome spaces.
Climate Modeling
Track evolving climate patterns in high-dimensional parameter space.
Earthquake Relief Prediction Market
See how spatial prediction markets work with a simulated earthquake disaster scenario.