Anomaly Detection with Joint Representation Learning of Content and Connection

TL;DR

This paper proposes a joint embedding approach combining content and network data to detect anomalous groups of users on social media, specifically applied to the 2019 Canadian Elections.

Contribution

It introduces an unsupervised method that jointly learns representations from user content and follower networks to identify suspicious user groups.

Findings

Detected densely-connected user groups engaging in political trolling.

Uncovered regional patterns of political manipulation.

Demonstrated effectiveness on real election-related social media data.

Abstract

Social media sites are becoming a key factor in politics. These platforms are easy to manipulate for the purpose of distorting information space to confuse and distract voters. Past works to identify disruptive patterns are mostly focused on analyzing the content of tweets. In this study, we jointly embed the information from both user posted content as well as a user's follower network, to detect groups of densely connected users in an unsupervised fashion. We then investigate these dense sub-blocks of users to flag anomalous behavior. In our experiments, we study the tweets related to the upcoming 2019 Canadian Elections, and observe a set of densely-connected users engaging in local politics in different provinces, and exhibiting troll-like behavior.