Twitter recently announced that it is going to begin using algorithms to help minimize abusive behavior on their site. The platform defines abusive behavior as violations of their TOS, or unsolicited tweeting at non-followers. The new system has been active for some weeks, but it was made public today. It's not just keyword hunting — it considers the relations between the users in question when figuring out if abuse is taking place.

Cracking down on abusive behavior

Penalties to accounts that were 'caught' by the algorithms will be punishments such as turning their tweets invisible to their followers, or being forced to associate an email address or phone to verify the account.

Advertisements
Advertisements

Ed Ho, #Twitter's vice president of engineering, spoke about the new algorithms and tools, saying that while the platform is meant to allow anyone to speak their mind no matter what their beliefs are, they will consider taking further action if an account keeps on engaging in abusive behavior.

Ho acknowledges that the approach might go through some growing pains, but assures that the company is actively working to improve and iterate on the new tools. As of now, there isn't a process to appeal any of the penalties yet.

Twitter's new tools to combat abuse

Other changes to the platform include the ability to not see notifications from accounts without a profile photo, or without a verified phone number of the email address or phone number, both usually hints that the account only exists to anonymously abuse users.

Advertisements

Users can also define for how long they want to filter out keywords, accounts and even conversations. All these new tools are similar to the verified users' notification settings and quality filter features, that have been in place for some time now.

The platform will also start sending in-app notifications about the state of reported tweets and accounts to those who flagged them. The company will ping the user when it has received their report, and then again when it has decided to take action. Before the changes, the notifications happened through email, a process that often left users confused about whether their report had even been received.

Twitter turning a new leaf

These decisions are part of an effort to help combat abusive behavior, as Twitter has received some criticism for not effectively handling it ever since its foundation.

Until not very long ago, the platform would not take action until users themselves reported suspected abuse. In fact, until fairly recently, Twitter required the victims of abusive behavior to make the reports themselves, refusing to take reports from friends, family or other third parties.