Measuring Node Similarity for the Collective Attention Flow Network.

Quantifying the similarity of nodes in collective attention flow network has an important theoretical and practical value. In this paper, we defined the generation time Rt, the influence radius Sr and the representation Vs (Rt, Sr) of the nodes in the collective attention flow network based on the optimization of Spatial Preferred Attachment (SPA) model. NID algorithm, based on the influence distance Sd that was calculated by the spatial norm, to measure the similarity of the nodes in the collective attention flow network was proposed. Experiments show that our algorithm not only accurately quantify the similarity of nodes in the collective attention flow network, but has a higher universality.