A Quality- Time Model of Heterogeneous Agents Measure for Crowd Intelligence

With the rapid progress of artificial intelligence (AI), it becomes increasingly important to measure the crowd intelligence of both machine and human under a same benchmark. However, previous studies mainly emphasize on task-specific skill or universal ability, ignoring the hybrid property of heterogeneous agents. In this work, we present a formal definition of crowd intelligence based on quality and time. The crowd intelligence is modeled as an aggregate of agent's multiple response-abilities on environment or external stimuli. We propose the concept of crowd intelligence quotient (CIQ) to describe the size of crowd intelligence, including human and machines. Furthermore, we propose a quality-time model of heterogeneous agents measure for crowd intelligence. For the hybrid property of heterogeneous agents, we use response-quality to describe the AI's characteristic of task and use response-time to describe the human's characteristic of ability. Finally, the actual data of AI and human is used to verify our proposed model. The experimental results demonstrate proposed QTM is better than traditional methods in both AI and human. Through the experimental results, we found that the CIQ of AI is about one-quarter of that of a normal adult human.