yfiua/user-role-prediction

This is the code for the paper "Predicting User Roles in Social Networks using Transfer Learning with Feature Transformation" in ICDMW'16.

user-role-prediction

This is the code for the paper "Predicting User Roles in Social Networks using Transfer Learning with Feature Transformation" in ICDMW'16.

Reproducing the Results

For reproduction Stu is recommended. If you have it installed you just need to perform:

stu

It will automatically download the wiki-talk datasets and do analysis. (You would need necessary Python packages installed.)

Datasets

Download the wiki-talk datasets using:

stu @dataset

Alternatively the datasets can be obtained at zenodo.org. If you want the up-to-date wiki-talk datasets you can follow the instruction here.

Reference

Please refer to the paper:

Sun J, Kunegis J, Staab S. Predicting User Roles in Social Networks Using Transfer Learning with Feature Transformation[C]// 2016 IEEE 16th International Conference on Data Mining Workshops (ICDMW). IEEE, 2016: 128-135.

@inproceedings{sun2016predicting,
  title={Predicting User Roles in Social Networks Using Transfer Learning with Feature Transformation},
  author={Sun, Jun and Kunegis, J{\'e}r{\^o}me and Staab, Steffen},
  booktitle={2016 IEEE 16th International Conference on Data Mining Workshops (ICDMW)},
  pages={128--135},
  year={2016},
  organization={IEEE}
}

user-role-prediction's Issues

Why these three?

Question: where are only three centrality measures are used here?
https://github.com/yfiua/user-role-prediction/blob/master/feature-aggr.py#L39