These scripts are meant to extract certain features from raw Twitter data describing Twitter users (tweets, profile info, as well as external data). Once the features are extracted, various forms of SVMs are trained, and logistic regressions are performed, to classify and rank the users. These operations are conducted on different subgroups of features. The details of the process are given in the below publications. The scripts were applied to the classification/ranking of Twitter users in terms of offline influence, based on the RepLab 2014 dataset.
- URL: https://github.com/CompNet/Influence
- Production date: 2014–2015
- Related publications:
- Jean-Valère Cossu, Vincent Labatut et Nicolas Dugué. « A Review of Features
for the Discrimination of Twitter Users : Application to the Prediction of Offline
Influence ». In : Social Network Analysis and Mining (SNAM) 6:25 (2016). DOI:
10.1007/s13278-016-0329-x. ⟨hal-01203171⟩ - Jean-Valère Cossu, Nicolas Dugué et Vincent Labatut. « Detecting Real-World
Influence Through Twitter ». In : 2nd European Network Intelligence Conference (ENIC).
Karlskrona, SE : IEEE Publishing, 2015, p. 83-90. DOI: 10.1109/ENIC.2015.20.
⟨hal-01164453⟩ (cite this publication if you use this software)
- Jean-Valère Cossu, Vincent Labatut et Nicolas Dugué. « A Review of Features