Google recently announced that it took some drastic action against ‘bad apps’ by removing a staggering 700,000 apps and 100,000 developer accounts in 2017. This is a sharp jump of 70 percent from the year before. Google claims to have used machine learning to detect apps with identifiers like, malware inappropriate content, and impersonation. "In fact, 99 percent of apps with abusive contents were identified and rejected before anyone could install them," the company claimed on Android Developers Blog.
Google stated that impersonation and ‘copycats’ are the most common reasons for removing certain apps from its Play store. In 2017, the company removed 250,000 apps that were reportedly impersonating bigger titles. The culprits went about this process using a host of deceptive techniques which included "confusable unicode characters or hiding impersonating app icons in a different locale."
Google’s Play Store also has strict rules against inappropriate content which is defined as any content that includes porn, extreme violence, hate and illicit activities. Google’s machine learning model helps identify the errant apps, which are then evaluated by human reviewers.
Finally, Google also red flags apps that can potentially harm devices. This includes malware, app involved in fraud and phishing. The Mountain View company claimed that the number of installs of Potentially Harmful Applications have reduced by 50 percent.
Andrew Ahn, Product Manager at Google Play, said that his team is committed to remove unwanted material from the app store. "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 defence. We take these extremely seriously, and will continue to innovate our capabilities to better detect and protect against abusive apps and the malicious actors behind them,” he said.