Revisiting device classification rules upon observation of new endpoint attributes

1. A method comprising:

using, by a device classification service, an initial device classification rule to label each of a set of endpoint devices in a network as being of a particular device type;

identifying, by a device classification service, a particular attribute exhibited by at least a portion of the set of endpoint devices, wherein the particular attribute was not previously used to generate the initial device classification rule;

generating, by the device classification service, one or more new device classification rules based in part on the particular attribute; and

switching, by the device classification service, from using the initial device classification rule to label endpoint devices in the network to using the one or more new device classification rules to label endpoint devices in the network.

2. The method as in claim 1 , further comprising:

maintaining, by the device classification service, a database of device attributes that were used to generate the initial device classification rule.

3. The method as in claim 1 , wherein switching from using the initial device classification rule to label endpoint devices in the network to using the one or more new device classification rules to label endpoint devices in the network comprises:

suggesting, via a user interface, the one or more new device classification rules; and

receiving, via the user interface, an acceptance of the one or more new device classification rules.

4. The method as in claim 3 , further comprising:

using the acceptance as feedback for a machine learning model that predicts an attribute relevancy score.

5. The method as in claim 1 , wherein generating the one or more new device classification rules based in part on the particular attribute comprises:

applying clustering to attributes associated with the set of endpoint devices, the attributes including the particular attribute and one more attributes on which the initial device classification rule was based.

6. The method as in claim 1 , further comprising:

computing, by the device classification service, a weighting for the particular attribute based on a fraction of the endpoint devices exhibiting the particular attribute to the set of endpoint devices, wherein the device classification service generates the one or more new device classification rules based on the weighting.

7. The method as in claim 1 , wherein the one or more new device classification rules comprise a device classification rule having fewer conditional clauses than that of the initial device classification rule.

8. The method as in claim 1 , wherein the one or more new device classification rules comprise a device classification rule that has a more granular device type label than that of the initial device classification rule.

9. The method as in claim 1 , further comprising:

sending an instruction to one or more networking devices in the network to increase collection of the particular attribute in the network.

10. The method as in claim 1 , wherein the initial device classification rule comprises one or more conditional clauses, each clause corresponding to a different device attribute.

11. An apparatus, comprising:

one or more network interfaces;

a processor coupled to the one or more network interfaces and configured to execute one or more processes; and

a memory configured to store a process that is executable by the processor, the process when executed configured to:

use an initial device classification rule to label each of a set of endpoint devices in a network as being of a particular device type;

identify a particular attribute exhibited by at least a portion of the set of endpoint devices, wherein the particular attribute was not previously used to generate the initial device classification rule;

generate one or more new device classification rules based in part on the particular attribute; and

switch from using the initial device classification rule to label endpoint devices in the network to using the one or more new device classification rules to label endpoint devices in the network.

12. The apparatus as in claim 11 , wherein the process when executed is further configured to:

maintain a database of device attributes that were used to generate the initial device classification rule.

13. The apparatus as in claim 11 , wherein the apparatus switches from using the initial device classification rule to label endpoint devices in the network to using the one or more new device classification rules to label endpoint devices in the network by:

suggesting, via a user interface, the one or more new device classification rules; and

receiving, via the user interface, an acceptance of the one or more new device classification rules.

14. The apparatus as in claim 13 , wherein the process when executed is further configured to:

use the acceptance as feedback for a machine learning model that predicts an attribute relevancy score.

15. The apparatus as in claim 11 , wherein the apparatus generates the one or more new device classification rules based in part on the particular attribute by:

applying clustering to attributes associated with the set of endpoint devices, the attributes including the particular attribute and one more attributes on which the initial device classification rule was based.

16. The apparatus as in claim 11 , wherein the process when executed is further configured to:

compute a weighting for the particular attribute based on a fraction of the endpoint devices exhibiting the particular attribute to the set of endpoint devices, wherein the device classification service generates the one or more new device classification rules based on the weighting.

17. The apparatus as in claim 11 , wherein the one or more new device classification rules comprise a device classification rule having fewer conditional clauses than that of the initial device classification rule.

18. The apparatus as in claim 11 , wherein the one or more new device classification rules comprise a device classification rule that has a more granular device type label than that of the initial device classification rule.

19. The apparatus as in claim 11 , wherein the process when executed is further configured to:

send an instruction to one or more networking devices in the network to increase collection of the particular attribute in the network.

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

using, by the device classification service, an initial device classification rule to label each of a set of endpoint devices in a network as being of a particular device type;

identifying, by a device classification service, a particular attribute exhibited by at least a portion of the set of endpoint devices, wherein the particular attribute was not previously used to generate the initial device classification rule;

generating, by the device classification service, one or more new device classification rules based in part on the particular attribute; and

switching, by the device classification service, from using the initial device classification rule to label endpoint devices in the network to using the one or more new device classification rules to label endpoint devices in the network.