“Google is using latest machine learning technique to eliminate bad apps from play store”
Google whose sole motto is “Don’t be Evil” does not want evil apps to capture the sole of your android phone. That’s why they will be using machine language to sort good apps from bad apps.
As we all knows that android is an open source operating systems, which means more security needed to protect your device. But that doesn’t mean Google is not doing enough to increase the security of your device especially from bad apps that may harm your device from viruses and hackers.
With the help of machine learning and AI, Google can now easily detect bad apps before you install them on your device.
Known as peer grouping, the company has described in detail how it’s using machine learning technique to sort good apps from bad apps to keep the Play Store a safe place to install apps that you are looking for on your android phone.
What is Peer Grouping?
Peer grouping compares data about apps that perform similar tasks, so they can identify the bad ones.
Imagine you are looking at a group of apps, the app that ask for permission to access your microphone, location, and phone book is probably the one that consider as bad app. Google’s new machine learning system will now flags that app into the category of harmful app automatically.
With machine learning technology Google can easily use peer grouping to scan apps that are being loaded on to the Play Store all together.
A range of metrics are used to sort good apps from bad apps like their description, their metadata (how big the file size is for example), and statistics like how many times they’ve been installed. As Google says a new peer group is created for each app, like “productivity” and “games” as a set categories to know the changing distinctions of the app world. Once grouped, we can easily pick out the bad apps “According to Martin Pelikan of Google’s security and privacy team, “We focus on signals that can negatively affect user privacy, such as permission requests that are not related to core app functionality, and the actual, observed behaviors,”. For example, a flashlight app might not need access to address book of the user or the precise hardware identifier of a user’s phone. The same might hold for many other apps, such as ‘mirror’ apps that turns on a device’s front-facing camera.”
Technique like Machine Learning and AI seems to be making a difference for Google as well for those who are installing apps. Recently at annual Android security review, Google confirmed that, the percentage of users who had installed harmful apps from the official Play Store fell from 0.15 percent in 2015 to 0.05 percent in 2016.