Figure 3 illustrates the process of community detection using algorithm NILP in the above example network when [alpha] = 2.
Algorithm NILP is different from other label propagation based algorithms.
At this point, NILP algorithm becomes the original label propagation algorithm LPA.
A large number of experiments show that, in most cases, the community-dividing results of the proposed algorithm NILP are optimal or near-optimal when [alpha] = 2.
Proof: The bijection proceeds by replacing each vertical step of the NILP with a 1 and each diagonal step with a 0 and vertically reflecting the array.
(In the NILP picture, each row has some number of vertical steps followed by some number of diagonal steps.)
It would seem reasonable to conjecture that the -1 of the ASM should correspond to all instances of a vertical step followed by a diagonal step as you go from left to right along a row of the NILP (or a 0 followed by a 1 as you go across a row of the boolean triangle).
Rejecting the null hypothesis of linearity for the NILP
deviations after accepting a long-run relationship between financial and insurance markets suggests that the NLIP is nonlinearly cointegrated with financial variables and that its dynamics are nonlinearly mean reverting toward equilibrium with a time-varying adjustment speed.
“The Summer of Service program was extremely beneficial to our organization,” says Tony Goodnough, Director of Finance and Administration for NILP
. “We were able to streamline our process and develop a system that actually saves us time and money.”