A Socio-Informatic Approach to Automated Account Classification on Social Media

Abstract

Automated accounts on social media have become increasingly problematic. We propose a key feature in combination with existing methods to improve machine learning algorithms for bot detection. We successfully improve classification performance through including the proposed feature.

arXiv commentInternational Conference on Social Media and Society

References in corpus (11)

• Emergence of scaling in random networks
• ranger: A Fast Implementation of Random Forests for High Dimensional Data in C++ and R
• Mixing patterns in networks
• The Rise of Social Bots
• The DARPA Twitter Bot Challenge
• Online Human-Bot Interactions: Detection, Estimation, and Characterization
• DNA-inspired online behavioral modeling and its application to spambot detection
• k-core decomposition: a tool for the visualization of large scale networks
• Social Turing Tests: Crowdsourcing Sybil Detection
• A Network Topology Approach to Bot Classification
• Deploying South African Social Honeypots on Twitter