| CPC G06F 40/20 (2020.01) [G06F 21/00 (2013.01); G06F 21/554 (2013.01); G06F 21/62 (2013.01); G06F 40/30 (2020.01); H04L 51/04 (2013.01); H04L 51/212 (2022.05); G06F 21/629 (2013.01); H04L 51/02 (2013.01)] | 17 Claims |
|
1. A computer-implemented method for performing security in a virtual environment, the method comprising:
generating, by at least one processor, a virtual environment configured to provide a virtual reality, wherein the virtual environment provides connectivity to a user device;
inputting, by the at least one processor to a machine learning model, consecutive segmented messages received from the user device communicating in the virtual environment in which the consecutive segmented messages combine to form a complete message that is displayed in the virtual environment, the consecutive segmented messages being from an avatar in the virtual environment, wherein the consecutive segmented messages forming the complete message are separate posts by the avatar in the virtual environment intended to bypass a security rule, wherein an individual one of the separate posts of the consecutive segmented messages from the avatar does not violate the security rule;
detecting that a combination of the consecutive segmented messages of the separate posts forms the complete message that violates the security rule using the machine learning model, wherein the machine learning model is trained to classify the complete message as violating or not violating the security rule, the machine learning model being trained on data comprising a plurality of complete messages formed by combining respective sets of consecutive segmented messages, the plurality of compete messages being labeled as violating or not violating the security rule;
in response to detecting the violation of the security rule, changing the virtual environment to obfuscate a display of the consecutive segmented messages forming the complete message such that a view of the complete message is no longer displayed in the virtual environment; and
in response to the violation of the security rule being by one or more user accounts that posted the consecutive segmented messages forming the complete message, changing the virtual environment to obfuscate each of the one or more user accounts that posted the consecutive segmented messages forming the complete message.
|