Algorithms for Adaptive Seed Minimization

Codes for the Trim, i.e., the adaptive algorithm for the adaptive SM problem that provides expected approximation ratio. Please refer to paper: https://arxiv.org/abs/1907.09668 for details.

Running Environment

Linux-based OS

Input requirements

Dataset files: dataset files are referred to $\textit{Tested-Dataset}$ repository.

Realizations files: please refer to $\textit{Tools}$ repository for source code to generate realizations.

Remark: please note that generated realization files are supposed to locate in the same folder of dataset file according to my implementation.

How to run

Running command

./algo -dataset path_to_dataset -model IC -epsilon ε -Q percentage -batch batch_size -seedfile filename -time time_number

Explain

--epsilon: an float number in range (0,1) to control the approximation error.

--Q: the percentage of nodes required to be activated.

--batch: the size of batch b selected each time.

--seedfile: the file records the k seed nodes selected.

--time: the number of the algorithm repeated.

For example:

./trim -dataset dataset/hep/ -model IC -epsilon 0.5 -Q 0.1 -batch 10 -seedfile seed -time 1

Datasets

Tested datasets can be downloaded in $\textit{Tested-Dataset}$ repository.

Remark

If there are any problems, please contact [email protected] / [email protected]. (https://sites.google.com/view/kekehuang/)