Twitter Now Automatically Detecting Abusive Accounts

Advertisement
Advertisement

Twitter's latest anti-abuse measure entails a set of solutions designed to automatically detect abusive accounts, the San Francisco-based social media giant announced on Wednesday. The company's Vice President of Engineering Ed Ho revealed how Twitter is currently in the process of leveraging its technology to not only sanction abusive content but also reduce the frequency at which users encounter online abuse on Twitter. Ho explained how Twitter already implemented some methods of detecting abusive accounts even if their problematic behavior hasn't been reported by other users. Once detected, the social media service limits their account functionality for a defined period of time. Ho revealed how there are many behavior types that could trigger this automated course of action and specifically mentioned accounts that often tweet at non-followers as an example. As there are numerous reasons for people on Twitter to tweet at users who they aren't following or who aren't following them, Twitter presumably placed some safety mechanism in place as this functionality could otherwise backfire relatively quickly.

Twitter's VP of Engineering also said how the social media platform will now automatically identify accounts that indulge in abusive behavioral patterns, but he has yet to elaborate on that feature. The aforementioned potential for false positives was also mentioned by Ho who admitted how all of the new measures Twitter is now rolling out are completely new and likely won't always work as intended. Due to that state of affairs, Twitter's anti-abuse measures will probably result in some mistakenly sanctioned people but the company will be reviewing all affected accounts and act accordingly, Ho said.

Other anti-abuse features that Twitter is now rolling out include new filtering options which allow users to have a more control over their notifications. For example, users will now be able to filter out notifications from users without a profile image or users who didn't verify their email addresses. Finally, the social media network also improved its mute feature which now allows people to mute other accounts directly from their Timeline and even mute them for a specified period of time. First screenshots depicting Twitter's latest anti-abuse features can be seen below.

Advertisement