Skip to content

uditarora/collusive-retweeters-TIST-2020

master
Switch branches/tags
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
 
 
 
 
 
 
 
 

collusive-retweeters-TIST-2020

Analyzing and Detecting Collusive Users Involved in Blackmarket Retweeting Activities (ACM TIST 2020)

This is the code and the dataset for the paper titled

Analyzing and Detecting Collusive Users Involved in Blackmarket Retweeting Activities. Udit Arora*, Hridoy Sankar Dutta*, Brihi Joshi, Aditya Chetan, Tanmoy Chakraborty

published at ACM Transactions on Intelligent Systems and Technology.

Quick Start

Requirements

  • Python 3.6+ To install the dependencies used in the code, you can use the requirements.txt file as follows -
pip install -r requirements.txt

Dataset

binary.csv and multi.csv contain the 50-dimensional user embeddings that give the best results for binary and multi-class classification of collusive/genuine users. Each line of the files represents a user, where the first 50 values represent the 50-dimension embedding, and the last value is the label.

user_views.tsv.gz contains the user views corresponding to each user. Each row of the file contains tab separated values corresponding to the following for each user -

  1. [label 1 1 1 1 1 1]: 'label' represents the annotated label for the user, and the six '1's represent the presence of the corresponding view for the user
  2. [AV1]: Tweet2Vec view
  3. [AV2]: SCoRe view
  4. [NV1]: Retweet network view
  5. [NV2]: Quote network view
  6. [NV3]: Follower network view
  7. [NV4]: Followee network view

Labels

  • label = 0: Genuine User
  • label = 1: Normal Customer
  • label = 2: Promotional Customer
  • label = 3: Bot

Running the code

Embedding generation

Use https://github.com/abenton/wgcca with user_views.tsv as input to generate the multi-view user embeddings using WGCCA.

Classification

Pick the type of dataset to be analyzed, and use the run_classifier method in classifiers.py to train a classifier and evaluate the performance.

License

Copyright (c) 2020 Udit Arora, Hridoy Sankar Dutta, Brihi Joshi, Aditya Chetan, Tanmoy Chakraborty

For license information, see LICENSE or http://mit-license.org

About

Analyzing and Detecting Collusive Users Involved in Blackmarket Retweeting Activities (ACM TIST 2020)

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages