Twitter are now stepping up its actions in combating users posting abusive or inappropriate content as it implemented a new measure to filter public tweets.
Techspot.com reported that Twitter is now injecting behavioral signals in order to identify users with troll-like behaviors that are committing abuse and posting contents that violate standards and policies.
According to VP of Trust and Safety Del Harvey and Director of Product Management for Health David Gasca, the signals are not externally visible as such identify whether a user did or did not confirm their e-mail address, if multiple accounts were created at the same time by the same user, users that repeatedly mention accounts that don’t follow them, or behavior indicating coordinated attack. Twitter also looks into accounts that are connected to violators and how they interact with each other.
Before implementing the new measure, Twitter previously used human review processes and policies and machine learning that determine how tweets in “communal” places are presented and organized through search and conversations.
According to Twitter, the improved measures showed improvement as abuse reports dropped by four percent from search and eight percent from conversations.
(Photo source: fortune.com)
