Parental monitoring of multiplayer game chat
A background application running on a child computing device periodically takes screenshots which are samples of what the child is viewing. If the child is playing a video game, text can be extracted from the screenshot within areas of the screen that chat is known to be displayed for the particular game. A cloud computing platform provides services for machine learning (ML) to analyze the chat text to ascertain a likelihood that the chat includes bullying or sexual predation. If it appears likely, the game is disabled and a notification is sent to a parental monitoring application along with a copy of the offensive chat. The parent can override the determination, and in this case, the user application running on the child computing device is re-enabled. The parent's action either to override the determination or to let it stand is then fed back to the ML model so that the ML model continues to learn the semantics of in-game chat based on the parental feedback.
1. A system for monitoring in-game chat, comprising:
a child computing device including a display, the child computing device capable of executing user applications and a background application periodically taking screenshots of the display;
a parent computing device in communication with the child computing device, the child computing device executing a parental monitoring application;
a device management server that establishes a master/slave relationship between the child computing device and the parent computing device; and
a cloud computing platform assisting the parental monitoring application that is configured to:
for the screenshots associated with an identified game, extract chat text from bounded areas of the screenshots, the bounded areas defined based on the identified game;
feed the extracted chat text to a machine learning model to evaluate one or more of bullying and predation;
determine a likelihood of one or more of bullying and predation based on the evaluation; and
notify the parent computing device, if the determined likelihood exceeds a predetermined threshold;
wherein the device management server causes the identified game executing on the child computing device to be disabled when the determined likelihood exceeds the predetermined threshold.
2. The system for monitoring in-game chat of claim 1 , wherein the identified game is provided to the cloud computing platform by the child computing device.
3. The system for monitoring in-game chat of claim 1 , wherein the identified game is determined by the cloud computing platform based on image analysis of one or more of the screenshots.
4. The system for monitoring in-game chat of claim 1 , wherein the machine learning includes one or more of a convolutional neural network (CNN), a recurrent neural network (RNN), a support vector machine (SVM), and a Bayesian Network.
5. The system for monitoring in-game chat of claim 1 , wherein the parent computing device allows a parent to override the determined likelihood.
6. The system for monitoring in-game chat of claim 5 , wherein the override is used to train the machine learning model.
7. The system for monitoring in-game chat of claim 5 , wherein the lack of an override is used to train the machine learning model.
8. The system for monitoring in-game chat of claim 5 , wherein, responsive to the override, the device management server causes the identified game to be re-enabled on the child computing device.