Abstract:
Crowd intelligence is an intelligent method for solving large-scale complex problems by aggregating group intelligence. Its idea is originated from the simulation of the intelligent behaviors of social biological swarms in nature. Through the division of labor, cooperation, coordination, and co-evolution, swarm creatures can emerge integrated intelligent behaviors. Swarms can complete complex tasks with a high degree of self-organization, self-ada-ptation and self-learning ability. Inspired by this, researchers use mathematics and computing tools to simulate the behavior of swarm intelligence, and develop a series of mechanisms and models based on the emergence and evolution of swarm intelligence. In recent years, with the development of the Internet, the collaborative and evolutionary phenomenon of human crowd intelligence based on the Internet of Things has further broadened the scope of crowd intelligence, presenting a broad application prospect, which also poses new challenges to the theoretical models and applications of group intelligence evolution. In 2017, "New Generation Artificial Intelligence Development Plan" clearly listed crowd intelligence as one of the important artificial intelligence theories and techniques to be deve-loped. In this report, the development of crowd intelligence has been studied from different perspectives: from biological swarms to multi-agent systems, and further to human crowds. The main research issues in crowd intelligence and evolutionary computing will be discussed in four aspects: theories and models of crowd intelligence, organization of crowds, crowd intelligence for collaborative decision-making, and the applications of crowd intelligence and evolutionary computing. The main research issues of crowd intelligence and evolutionary computation is summarized in this report. The reviews have also been made for the latest studies in China and abroad. Finally, some discussion has also been made for future trends and potential scientific issues of this research field.