Is this correct to sample a node under LT model? - Githubissues

snowgy / Influence_Maximization

Simple and Fast State-of-Art Implementation of Influence Maximization Problem

47 stars 10 forks source link

Is this correct to sample a node under LT model? #1

Open comingx opened 5 years ago

comingx commented 5 years ago

https://github.com/snowgy/Influence_Maximization/blob/941ebbb765631beda82b267eb2f1c3c79d377c7f/IMP.py#L227

This line shows that it randomly choose an incoming node. But the author's C++ code goes like: https://github.com/CorentinDieudonne/IMM/blob/1a8f36a8887653d55ae203224b956e61b23c7425/TIM%2B-Release/src/infgraph.h#L146

Juan-Chen45 commented 3 years ago

The LT sampling model is wrong, please refer to Revisiting the Stop-and-Stare Algorithms for Influence Maximization, http://www.vldb.org/pvldb/vol10/p913-Huang.pdf, see Appendix A