SoylentNews

tonyPick writes:

Spotted on HackerNews is a link to a paper on Automated Crowdturfing Attacks and Defenses in Online Review Systems:

Malicious crowdsourcing forums are gaining traction as sources of spreading misinformation online, but are limited by the costs of hiring and managing human workers. In this paper, we identify a new class of attacks that leverage deep learning language models (Recurrent Neural Networks or RNNs) to automate the generation of fake online reviews for products and services. Not only are these attacks cheap and therefore more scalable, but they can control rate of content output to eliminate the signature burstiness that makes crowdsourced campaigns easy to detect.

The paper, available from the arXiv link, contains the details of the attack which the paper notes "are largely indistinguishable from real reviews to human readers", and suggests defensive mechanisms based on "statistically detectable variations in the character-level distribution of machine-generated reviews".

Original Submission