We make use of the [mutually reinforcing] relationship between … [the authority weight and the hub weight]. … [p. 612] Now, to find the desired "equilibrium" values … [equations for updating hub and authority weights] in an alternating fashion, and see whether a fixed point … [p. 613] [The algorithm converges to the principal eigenvectors of the weighting matrices.] … [p. 614] In our experiments, we find that the … c largest coordinates in each vector to become …
Google-1
Google-2
Copyright clearance needed for quotation.