Iran-based horizontal marketplace Divar has begun to use AI to identify users who violate its terms of use and restrict their activity on the site.

If a user is flagged and added to a list of those identified as having a high probability of violating Divar’s terms due to suspicious activities or user reports, they can still view ads but cannot post any ads themselves, see the contact details of advertisers or use the chat feature, according to the company’s blog post.

However, the user’s rights are restored if they pass an authentication process that consists of a national ID check and photo and video verification. This process is similar to that required to open an online bank account in the country.

The AI has been trained to distinguish between fake and genuine user reports: “In order to increase the security of user transactions and prevent the activities of fraudsters, Divar has developed its artificial intelligence model to identify people with suspicious behavior of violation and fraud. This model is now operating on the platform, and its results have been highly accurate,” said Divar.

“This artificial intelligence model, taking into account different characteristics, detects users who behave suspiciously and classifies them into categories with different probabilities of having committed a violation, based on their activity. These users face various punishments or restrictions based on the risk level of their activity.”

This is the latest in a series of security measures introduced by Divar, which was founded in 2012. It has already introduced a video authentication service and enabled user feedback on property ads.

Print Friendly, PDF & Email

Related Articles