Exploit prediction based on machine learning

A system comprising:

• storage media;
• one or more processors;
• and one or more programs stored in the storage media and configured for execution by the one or more processors, the one or more programs comprising instructions for:
• causing generation of a prediction model based on selecting training data comprising one or more prevalence features for each software vulnerability of a first plurality of software vulnerabilities, wherein the prediction model enables determining whether an exploit that has yet to be developed is likely to be used in an attack;
• causing application of the prediction model to input data comprising one or more prevalence features for each software vulnerability of a second plurality of software vulnerabilities, wherein each software vulnerability of the second plurality of software vulnerabilities has yet to have an exploit developed for said each software vulnerability;
• receiving, based on the application of the prediction model to the input data, first output data that indicates, for each software vulnerability of the second plurality of software vulnerabilities, a prediction of whether an exploit that is predicted to be developed for said each software vulnerability is also predicted to be used in an attack.

A method comprising:

• causing generation of a first prediction model based on selecting first training data comprising a prevalence feature for each software vulnerability of a first plurality of software vulnerabilities, wherein the first prediction model enables determining whether an exploit is likely to be developed for a particular software vulnerability;
• causing application of the first prediction model to first input data comprising the prevalence feature for each software vulnerability of a second plurality of software vulnerabilities, wherein each software vulnerability of the second plurality of software vulnerabilities has yet to have an exploit developed for said each software vulnerability;
• receiving, based on the application of the first prediction model to the first input data, first output data that indicates, for each software vulnerability of the second plurality of software vulnerabilities, a prediction of whether an exploit will be developed for said each software vulnerability;
• causing generation of a second prediction model based on selecting second training data corresponding to a subset of the first plurality of software vulnerabilities, wherein each software vulnerability in the subset of the first plurality of software vulnerabilities already has an exploit developed for said each software vulnerability, wherein the second prediction model enables determining whether an exploit that has yet to be developed is likely to be used in an attack;
• causing application of the second prediction model to second input data corresponding to a subset of the second plurality of software vulnerabilities, wherein each software vulnerability in the subset of the second plurality of software vulnerabilities is predicted to have an exploit developed for said each software vulnerability;
• receiving, based on the application of the second prediction model to the second input data, second output data that indicates, for each software vulnerability in the subset of the second plurality of software vulnerabilities, a prediction of whether an exploit that is predicted to be likely to be developed for said each software vulnerability is also predicted to be likely to be used in an attack;
• wherein the method is performed using one or more computing devices.