Timezone: »

 
Curating the Twitter Election Integrity Datasets forBetter Online Troll Characterization
Albert Orozco Camacho · Reihaneh Rabbany

Tue Dec 07 01:26 PM -- 01:31 PM (PST) @
in LatinX in AI (LXAI) Research @ NeurIPS 2021 »

In modern days, social media platforms provide accessible channels for the inter-action and immediate reflection of the most important events happening around the world. In this paper, we, firstly, present a curated set of datasets whose origin stem from the Twitter’s Information Operations efforts. More notably, these accounts, which have been already suspended, provide a notion of how state-backed human trolls operate.Secondly, we present detailed analyses of how these behaviours vary over time,and motivate its use and abstraction in the context of deep representation learning:for instance, to learn and, potentially track, troll behaviour. We present baselinesf or such tasks and highlight the differences there may exist within the literature.Finally, we utilize the representations learned for behaviour prediction to classify trolls from"real"users, using a sample of non-suspended active accounts.

Author Information

Albert Orozco Camacho (McGill University / Mila - Québec AI Insititute)

My name’s Albert, I am originally from Guadalajara, México and I enjoy all things artificial intelligence. 😎 Currently, I am a Research Master’s student at Mila - Québec AI Institute and the Reasoning and Learning Lab of McGill University under Prof Reihaneh Rabbany’s supervision. My current research interests rely in the intersection of Network Science and Natural Language Processing. I have a broad spectrum of interests in AI: from the theoretical foundations of deep learning to applications regarding NLP, chatbots, and social networks. My current goals are directed towards enhancing how humans and machines communicate and understand themselves, as well as providing elegant models for such tasks.

Reihaneh Rabbany (McGill University)

More from the Same Authors