A Multi-Agent Approach to Binary Classification Using Swarm Intelligence

Wisdom-of-Crowds-Bots (WoC-Bots) are simple, modular agents working together in a multi-agent environment to collectively make binary predictions. The represent knowledge-diverse crowd, with each agent trained on subset of available information. A honey-bee-derived swarm aggregation mechanism is used elicit collective prediction an associated confidence value from the agents. Due their design, WoC-Bots can be distributed across multiple hardware nodes, include new features without re-training existing agents, and incorporate predictions other sources, thus improving overall predictive accuracy system. In addition these advantages, we demonstrate that competitive top classification methods three datasets apply our system real-world sports betting problem, producing consistent return investment 1 January 2021 through 15 November 2022 most major sports.