A Social Diffusion Model with an Application on Election Simulation

Algorithm 2

Preference deployment method.

𝑆: social-networks; 𝑃: preference-matrix

𝐾: community-number

𝑀[𝑖]: the 𝑖th largest community

𝐶[𝑖]: the 𝑖th largest cluster

(𝑀[],𝐾)=𝐶𝑜𝑚𝑚𝑢𝑛𝑖𝑡𝑦𝐷𝑒𝑡𝑒𝑐𝑡𝑖𝑜𝑛(𝑆)

𝐶[]=𝐾𝑚𝑒𝑎𝑛𝑠𝐹𝑜𝑟𝑃𝑟𝑒𝑓𝑒𝑟𝑒𝑛𝑐𝑒(𝑃,𝐾)

for  𝑖=1  to  𝐾  do

 if  𝑠𝑖𝑧𝑒(𝐶[𝑖])=𝑠𝑖𝑧𝑒(𝑀[𝑖])   then

      next

 else if  𝑠𝑖𝑧𝑒(𝐶[𝑖])<𝑠𝑖𝑧𝑒(𝑀[𝑖])   then

      add the unassigned preferences which is outside

   𝐶[𝑖] but closest to 𝑐𝑒𝑛𝑡𝑒𝑟(𝐶([𝑖]) into 𝐶[𝑖] to

      match 𝑠𝑖𝑧𝑒(𝑀[𝑖])

 else

      remove the preferences most away from

      𝑐𝑒𝑛𝑡𝑒𝑟(𝐶[𝑖]) to the closest unassigned neighbor

      cluster

 end if

 randomly assign the preferences in 𝐶[𝑖] to nodes in

 𝑀[𝑖]

end for

return the pairs of (node, preference)