Twitter Fades Account "Spam" and "Bot" with ML

in #twitter6 years ago

Twitter Indonesia spends its latest attempt to eradicate fake accounts, spam, and bots, which disrupt users' convenience in sharing information. If previously only based on reports, now Twitter began to rely on machine learning technology (ML).

In his official blog, Twitter said that ML owned is able to identify and eradicate the network of spam accounts automatically. That way, Twitter can prevent the ignorant who want to manipulate the conversation.

In trials during May 2018, the ML ML system found and cracked down on more than 9.9 million spam accounts per week. That number rose significantly compared to 6.4 million accounts in 2017 ago.

"Due to technological improvements, we can now remove 214 percent more accounts that violate our spam policies from year to year," it says on the official Twitter blog.

"At the same time, the average number of spam reports received by our system continues to decline, from 25,000 per day in March and now only 17,000 per day," as quoted by KompasTekno, Friday (29/6/2018).

image

First, reducing the visibility of suspicious account to put it to the list of "read-line". Is account suspicious is account actions coordinated and mass. The account will immediately removed from the column follower people and a variety of involvement to qualify checking, such as confirmation phone number. Account soon restored third suspicion Twitter proven wrong. Second, improve the registration process. Twitter ask for a new account to confirm your email address or phone number when register. So, spam account will be more difficult to register. Third, audit the accounts of the existing to see whether or not there signs registration automatically. It is to ensure that each account through some security check automatic simple. Thanks to this effort, Twitter claims over 5,000 registration spammy have prevented at the beginning.

Finally, the expansion of detection system behavior dangerous. There are some process automated when the system kiss something suspicious. It is to speed up all the process the eradication. With its efforts, Twitter hope microblogging his can be a place to discuss healthy, and sharing of information is accurate and relevant.

Sort:  

Congratulations @glawa! You received a personal award!

Happy Birthday! - You are on the Steem blockchain for 1 year!

Click here to view your Board

Support SteemitBoard's project! Vote for its witness and get one more award!

Congratulations @glawa! You received a personal award!

Happy Birthday! - You are on the Steem blockchain for 2 years!

You can view your badges on your Steem Board and compare to others on the Steem Ranking

Vote for @Steemitboard as a witness to get one more award and increased upvotes!