Google's cleaning up the Play Store, using machine learning to remove 700,000 bad apps

From PC World: Earlier this month, Google removed some 60 games from the Play Store after it was uncovered that the so-called Adult Swine bug was delivering inappropriate and malicious ads to kids. According to Google, that’s just a drop in the bucket compared to the 700,000 it booted from the Play Store in 2017.

In a blog post titled, “How we fought bad apps and malicious developers in 2017,” Google spotlights the numerous ways it has stayed ahead of bad Play Store Apps over the past 12 months. In addition to the 700,000 apps it took down in 2017—a 70-percent increase over 2016’s numbers—Google also says it was able to identify and root them out much quicker: The company boasts that 99 percent of apps “with abusive contents” were tossed from the Play Store before anyone could install them. Google also says it has developed new detection techniques to weed out repeat offenders and identify “abusive developer networks at scale” to stop them from simply submitting another set of bad apps.

According to Google, the biggest reason for the jump in removed applications was machine learning. Google used its AI-powered engine to help “detect abusive app content and behaviors—such as impersonation, inappropriate content, or malware” which then helped the human reviewers detect problematic apps.

However, Google admits there is more work to be done, as evidenced by the Adult Swine bug: “Despite the new and enhanced detection capabilities that led to a record-high takedowns of bad apps and malicious developers, we know a few still manage to evade and trick our layers of defense.”

View: Article @ Source Site