Abstract: Based on the method of social network analysis, taking for Shanxi province for example, this paper studies the network characteristics of the whole network such as density and structure hole from the perspective of talent aggregation effect, and divides the whole network analysis into research and application. The results show that the densities of government, universities and enterprises are higher the whole network density, and the low density of the scientific research institutes network weaken the relationship between innovation subjects; besides these, this results show that the these two types of the collaborative innovation network appeared to vertical and surround distribution without enough talent aggregation cohesion. The research conclusions can be for the central and western regions of the local government, universities, research institutes and enterprises to provide the references for police making in promoting talent aggregation, improving talent aggregation effect, and promoting innovation-driven development.

Key words:  talent aggregation effect, region, collaborative innovation network, social network analysis