Network security system and method for preemptively identifying or remediating security vulnerabilities

A non-transitory computer readable storage medium storing network security analysis and vulnerability prediction program instructions for causing a network security appliance to analyze network security data from diverse data sources to predict a vulnerability at a node in a computer network for remediation, the program instructions comprising the steps of:

• receiving network security data from a plurality of data sources;
• extracting feature vectors from the received network security data;
• applying a machine learning model to the extracted feature vectors to generate a weighted vulnerability value;
• predicting a computing resource vulnerability at a node in the computer network using the weighted vulnerability value;
• evaluating the predicted computing resource vulnerability;
• determining whether the predicted computing resource vulnerability includes a false positive prediction;
• responsive to a false positive prediction, optimizing weighting in the machine learning model based on the extracted feature vectors;
• and
• transmitting the predicted computing resource vulnerability to a computing device which is configured to remediate the predicted computing resource vulnerability.

A network security system for analyzing network security data from diverse data sources to predict and remediate a vulnerability at a node in a computer network, the system comprising:

• a plurality of network security data sources that monitor and log activities and events in the computer network, the network security data sources including two or more of a network firewall, a router, an intrusion prevention system, a vulnerability scanner, a dynamic asset discovery system, and a static asset inventory manager;
• and
• a network security appliance that
• receives network security data from the plurality of data sources,
• extracts feature vectors from the received network security data,
• applies a machine learning model to the extracted feature vectors to generate a weighted vulnerability value,
• predicts a computing resource vulnerability at a node in the computer network using the weighted vulnerability value,
• evaluates the predicted computing resource vulnerability,
• determines whether the predicted computing resource vulnerability includes a false positive prediction,
• responsive to a false positive prediction, optimizes weighting in the machine learning model based on the extracted feature vectors, and
• transmits the predicted computing resource vulnerability to a computing device which is configured to remediate the predicted computing resource vulnerability.