Supporting files.

Fig A, The example to illustrate the basic idea of embedding algorithm. Fig B, The positions’ variation over the computation iterations. Fig C, The normalized cumulative curves of attention flows (brown diamond), attention dissipations (blue squares) and the number of websites (green circles) along radius on March 10, 2007. The inset shows the density curves of the three quantities and the derivatives to R of the three fitted “S” curves. Fig D, The Lorenze-liked curves and the GINI-liked coefficients among cumulative attention flows, attention dissipations, and the number of websites along the radius on March 10, 2007. The green nodes in the two sub-figures on the left represent the attention flows or dissipations of the 20% websites in the core. The insets show the log-log plots of the focal variable pairs. Fig E, The normalized cumulative curves of attention flows (brown diamond), attention dissipations (blue squares) and the number of websites (green circles) along radius on September 10, 2007. The inset shows the density curves of the three quantities and the derivatives to R of the three fitted “S” curves. Fig F, The Lorenze-liked curves and the GINI-liked coefficients among cumulative attention flows, attention dissipations, and the number of websites along the radius on September 10, 2007. The green nodes in the two sub-figures on the left represent the attention flows or dissipations of the 20% websites in the core. The insets show the log-log plots of the focal variable pairs. Fig G, The normalized cumulative curves of attention flows (brown diamond), attention dissipations (blue squares) and the number of websites (green circles) along radius on February 10, 2008. The inset shows the density curves of the three quantities and the derivatives to R of the three fitted “S” curves. Fig H, The Lorenze-liked curves and the GINI-liked coefficients among cumulative attention flows, attention dissipations, and the number of websites along the radius on February 10, 2008. The green nodes in the two sub-figures on the left represent the attention flows or dissipations of the 20% websites in the core. The insets show the log-log plots of the focal variable pairs. Fig I, The special case in calculating GINI-liked coefficients. Fig J, The distribution of attention flows on October 10, 2006. Fig K, The distribution of attention dissipations on October 10, 2006. Fig L, The distribution of websites on October 10, 2006. (PDF)