Download the dataset(DS-1) using this link:
https://drive.google.com/file/d/14lKOu5i-d3SrDuhe7iRzS0RSEpRTMHjh/view?usp=sharing
The data folder contains actual bots in 2016 and 2018 datasets of DS-1 and Preprocessed Dataset 2.
The code folder contains the following:
DS_2016_tier_approach.ipynb - Contains code related to calculating the bots based on tier 1 and tier 2 approach along with evaluation metrics code.This file requires input intermediate results from after calculating Threshold 1 and this file yields output total bots predicted from tier 1 approach , total bots predicted from tier 2 approach , total bots correctly predicted.
Ds_2_preprocess.ipynb - This file takes raw DS-2 datafile and converts it to coornet datafile format.
coor_net.py - This file contains code related to calculating Threshold-1(coordination interval) and code to construct highly coordinated graph.
network_mine.py - This file contains code which models the given dataset to community detection algorithms like k-clique and max modularity and clustering algorithms like DBSCAN.
dataset2_second_threshold_calculation.py - This file contains code related to calculating the second threshold in DS-2 and also code for most boosted accounts and most boosted domains.
Evaluations folder contains predicted botnames for Ds_1-2016, Ds_1-2018, Ds-2, also the visualizatons folder that contains GEPHI files and most boosted accounts and domains for datasets.