Preserving privacy in exporting device classification rules from on-premise systems

An apparatus, comprising:

• one or more network interfaces to communicate with one or more networks;
• a processor coupled to the network interfaces and configured to execute one or more processes;
• and
• a memory configured to store a process executable by the processor, the process when executed configured to:
• obtain data indicative of a device classification rule, a device type label associated with the rule, and a set of positive and negative feature vectors used to create the rule;
• replace similar feature vectors in the set of positive and negative feature vectors with a single feature vector, to form a reduced set of feature vectors;
• apply differential privacy to the reduced set of feature vectors;
• and
• send a digest to a cloud service,
• wherein the digest comprises the device classification rule, the device type label, and the reduced set of feature vectors to which differential privacy was applied, and
• wherein the cloud service uses the digest to train a machine learning-based device classifier.

A tangible, non-transitory, computer-readable medium storing program instructions that cause a device to execute a process comprising:

• obtaining, by a device in a network, data indicative of a device classification rule, a device type label associated with the rule, and a set of positive and negative feature vectors used to create the rule;
• replacing, by the device, similar feature vectors in the set of positive and negative feature vectors with a single feature vector, to form a reduced set of feature vectors;
• applying, by the device, differential privacy to the reduced set of feature vectors;
• and
• sending, by the device, a digest to a cloud service,
• wherein the digest comprises the device classification rule, the device type label, and the reduced set of feature vectors to which differential privacy was applied, and
• wherein the cloud service uses the digest to train a machine learning-based device classifier.