I'd be curious to see some math on this. The top ranking is probably following a power law, such as the most cited papers etc. etc. Now, what other weighting al

I'd be curious to see some math on this. The top ranking is probably following a power law, such as the most cited papers etc. etc. Now, what other weighting algorithm would you suggest? how to find the most interesting nodes? did someone ever published a proper study? What clustering will be predicted?