While there is still much work to be done, people are experiencing significantly less abuse on Twitter now than they were six months ago, the micro-blogging website has said.
In an update on its efforts to improve user safety and anti-abuse measures on its platform, Twitter said its team is now taking action on 10 times the number of abusive accounts every day compared to the same time last year.
"We also now limit account functionality or place suspensions on thousands more abusive accounts each day," Ed Ho, General Manager, Consumer Product and Engineering at Twitter, said in a blog post late Thursday.
There are repeat offenders who create new accounts after being suspended for violations.
"Our new systems, in the last four months alone, have removed twice the number of these types of accounts. Beyond the technology, our teams are continuing to review content daily and improve how we enforce our policies," Ho posted.
In its bid to reduce abuse, the micro-blogging website recently introduced new optional filters to prevent a user from being notified when certain accounts tweet at them.
Twitter is augmenting the new filters with machine learning.
"Accounts that demonstrate abusive behaviour are now limited for a time, and told why. Accounts that we put into this period of limited functionality generate 25 per cent fewer abuse reports, and approximately 65 per cent of these accounts are in this state just once," the post further read.
"Our ‘Quality Filter' has led to fewer unwanted interactions: blocks after @mentions from people you don't follow are down 40 per cent. These numbers will vary, but the approach is having a positive impact," Ho said.
Ho added that Twitter would continue to collaborate with its Trust and Safety Council, industry experts and academics to solicit ongoing feedback.