18158623. Network Anomaly Detection simplified abstract (GOOGLE LLC)

Network Anomaly Detection

Organization Name

GOOGLE LLC

Inventor(s)

Mikhal Shemer of Tel Aviv (IL)

Roee Engelberg of Tel Aviv (IL)

Yonit Tova Halperin Worzel of Tel Aviv (IL)

Alex Gontmakher of Holon (IL)

Alexander Goldshtein of Tel Aviv (IL)

Gal Elidan of Modiin (IL)

Benjamin Dov Kessler of Jerusalem (IL)

Network Anomaly Detection - A simplified explanation of the abstract

This abstract first appeared for US patent application 18158623 titled 'Network Anomaly Detection

Simplified Explanation

The patent application describes a technology for monitoring and detecting network anomalies in a cloud network using machine learning techniques. Here are the key points:

• Cloud networks are complex environments with numerous users and virtual machines.
• Virtual machines can have unknown and complex behavior.
• The technology uses machine learning to monitor the cloud network and detect anomalies.
• User feedback is incorporated to modify and adapt the models for specific use cases, virtual machine types, and users.

Potential Applications

This technology has potential applications in various areas, including:

• Cloud service providers can use it to enhance the security and stability of their networks.
• Enterprises using cloud services can benefit from improved anomaly detection and network monitoring.
• IT teams can utilize this technology to identify and address potential issues in their cloud infrastructure.

Problems Solved

The technology addresses the following problems:

• Complex behavior of virtual machines in cloud networks makes it challenging to detect anomalies.
• Traditional monitoring methods may not be effective in identifying network anomalies in a cloud environment.
• Lack of adaptability to different use cases and user requirements hinders accurate anomaly detection.

Benefits

The technology offers several benefits:

• Improved detection of network anomalies in cloud environments.
• Enhanced security and stability of cloud networks.
• Ability to adapt and customize the anomaly detection models based on user feedback.
• Better understanding and management of complex virtual machine behavior.

Original Abstract Submitted

A cloud network is a complex environment in which hundreds and thousands of users or entities can each host, create, modify, and develop multiple virtual machines. Each virtual machine can have complex behavior unknown to the provider or maintainer of the cloud. Technologies disclosed include methods, systems, and apparatuses to monitor the complex environment to detect network anomalies using machine learning techniques. In addition, techniques to modify and adapt to user feedback are provided allowing the developed models to be tuned for specific use cases, virtual machine types, and users.