Twitter to Combat Trolls With Algorithm Changes
New algorithm changes introduced by the platform will combat the problem of social media “trolls” by exiling them to the lower rungs of Twitter threads.
Twitter says it will begin taking active steps to combat online trolls on its platform by making them less visible, claiming that accounts such as these distort and detract from the public conversation.
For several years now, the platform has been trying to tackle the seemingly endless torrent of abuse and harassment that occurs on it daily. This latest move is another step toward eliminating unwanted comments and statements by individuals who dedicate their time to disrupting debate and discourse.
Combating the Problem
The company says it will begin to look at behavioural signals to improve the health of discussion and debate on the platform. In a statement on the its blog, Del Harvey, Vice President of Trust and Safety and David Gasca, Product Manager for Health, said:
“Now, we’re tackling issues of behaviours that distort and detract from the public conversation in those areas by integrating new behavioural signals into how Tweets are presented.”
“By using new tools to address this conduct from a behavioural perspective, we’re able to improve the health of the conversation, and everyone’s experience on Twitter, without waiting for people who use Twitter to report potential issues to us.”
Tweets to be Ranked
The upcoming algorithm changes will see tweets in conversations and search be ranked based on a far greater variety of data. The ranking will take into account things such as the number of accounts registered to that user and whether or not a specific tweet prompted other users to block the account.
Tweets that are determined to be toxic or suspicious aren’t deleted, however, they are moved down the pecking order in the tweet conversation into the show more replies section. This will result in fewer users noticing the questionable posts.
Some examples of suspicious or toxic behaviour highlighted by Twitter include:
- One person registering several accounts at the same time
- A person repeatedly tweeting at people who do not follow them
- Taking part in coordinated attacks on another person
Twitter said that people contributing to conversation in a positive and healthy way will benefit from the change. Good tweets will be ranked higher up the list of visible replies on a post. This method of reward, could ultimately encourage users to interact and contribute in a more civilised manner.
According to Twitter the results so far have been encouraging, so far there has been a 4% drop in abuse reports from search and 8% fewer abuse reports from conversations. However, the task is far from done, the company says. This process makes up only one part of their efforts to improve the health of online conversation, the company says that this will be a long journey of learning and adaption and it is likely there will be mistakes along the way.
A Period of Change
This move by Twitter comes as another social network, Facebook, continues to improve user safety and cut down on harassment. Facebook is currently in the midst of a serious investigation into the use – and misuse – of data by thousands of apps over the past several years.
Social media companies are becoming increasingly aware of the potentially toxic activity that goes on across all platforms and, under increased pressure from users, are likely to continue improving their safety and reporting features to prevent abuse.