Google, announced that Google Play store in 2017 acted against the bad apps by removing apps and developer accounts from the platform. Play store is the home to millions of apps, out of which, some are good, while some are bad apps. In Google’s language, bad apps are the ones that don’t follow Play store’s policy. The tech giant said that it has used machine learning to identify bad apps with identifiers like impersonation, inappropriate content, and malware, which led them to eliminate over 700,000 apps and 100,000 developers in 2017, which google claims to be a 70 percent jump from the previous year.
The tech giant claim that they have not only removed the bad apps from Play store but have also identified and taken actions against them before anyone could install them. Google says that they have identified around 99 per cent of apps with abusive contents and rejected before anyone could install them. Stating the reason on how they managed to do this through new machine learning models and techniques. Google also says that they have developed some new detection models and techniques through which they will be able to identify repeated offenders and abusive developer networks. These techniques have helped Google take down around 100,000 bad developers in 2017 and has made it more difficult for them to create new accounts and flood Play store with bad apps.
Let us now look into the techniques that Google considered to identify a bad app on Play store in order to take necessary actions against them:
Impersonation or Copycats: Google, on its Blog, said that Copycats or impersonating famous apps is one of the most common violations that lead to removing apps from Google Play. The tech giant claims that in 2017 alone, they have removed as much as 250,000 apps that were caught impersonating big titles. The impersonators carried out this practice through deceptive methods such as confusable Unicode characters or hiding impersonating app icons in a different locale.
Inappropriate Content: Google Play does not allow any kind of inappropriate content. According to the company Inappropriate content, includes pornography, extreme violence, illegal activities, and hate. Google claims that its advanced machine learning models help quickly sift through app submissions and flag them for inappropriate content. Google has reportedly removed tens of thousands of apps with such inappropriate content in the past year.
Malware: Finally, another important reason for Google that lead in removing Apps is the presence of Potentially Harmful Applications (PHAs) that can cause harm to the device and the users. Apps that conduct SMS fraud, act as trojans, or phishing user’s information are part of this list. The tech giant claims PHAs are currently small in volume but research to remove them is being heavily invested in. The company adds upon by saying that with the launch of Google Play Protect, which is Google’s malware scanning feature at I/O 2017, the annual PHA installs have apparently reduced to 50 percent year on year.
However, developers of these bad apps come out with new ways to crack these new protection features, for which Andrew Ahn, Product Manager at Google Play, said, “Despite the new and enhanced detection capabilities that led to some record-high takedowns of bad apps and malicious developers, we know a few still manage to evade and trick our layers of defence. We take these extremely serious and will continue to innovate our capabilities to better detect and protect against abusive apps and the malicious actors behind them. We are committed to make Google Play the most trusted and safe app store in the world”.