Time disambiguation for natural language queries in network monitoring systems
A device may obtain a query from a user interface regarding a computer network. The device may retrieve, based on the query, schema metadata from a database. The device may resolve any time ambiguity in the query using the schema metadata. The device may provide an answer to the query to the user interface, after resolving any time ambiguity in the query using the schema metadata.
1 . A method comprising:
obtaining, by a device, a query from a user interface regarding a computer network;
retrieving, by the device and based on the query, schema metadata from a database;
resolving, by the device, any time ambiguity in the query using the schema metadata; and
providing, by the device, an answer to the query to the user interface, after resolving any time ambiguity in the query using the schema metadata.
2 . The method as in claim 1 , wherein resolving any time ambiguity in the query comprises:
determining a desired time window associated with the query, wherein the query from the user interface does not specify any time window.
3 . The method as in claim 1 , wherein the device uses a large language model (LLM) to resolve any time ambiguity in the query using the schema metadata.
4 . The method as in claim 1 , wherein resolving any time ambiguity in the query using the schema metadata comprises:
sending, by the device and to the user interface, a request for a user to resolve any time ambiguity in the query, based on the schema metadata.
5 . The method as in claim 1 , wherein the schema metadata is for time series data in the database captured from the computer network.
6 . The method as in claim 5 , wherein the schema metadata is indicative of at least one of: a number of records per time unit for the time series data or a maximum time window for the time series data.
7 . The method as in claim 1 , wherein the query asks a status of a particular networking device in the computer network.
8 . The method as in claim 1 , wherein the device resolves any time ambiguity in the query based in part on contextual data associated with a user that issued the query via the user interface.
9 . The method as in claim 1 , further comprising:
generating, by the device, the answer to the query in part by translating the query into an application programming interface (API) call, after resolving any time ambiguity in the query using the schema metadata.
10 . The method as in claim 1 , further comprising:
generating, by the device, the answer to the query in part by translating the query into an SQL statement, after resolving any time ambiguity in the query using the schema metadata.
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:
obtain a query from a user interface regarding a computer network;
retrieve, based on the query, schema metadata from a database;
resolve any time ambiguity in the query using the schema metadata; and
provide an answer to the query to the user interface, after resolving any time ambiguity in the query using the schema metadata.
12 . The apparatus as in claim 11 , wherein the apparatus resolves any time ambiguity in the query by:
determining a desired time window associated with the query.
13 . The apparatus as in claim 11 , wherein the apparatus uses a large language model (LLM) to resolve any time ambiguity in the query using the schema metadata.
14 . The apparatus as in claim 11 , wherein the apparatus resolves any time ambiguity in the query by:
sending, to the user interface, a request for a user to resolve any time ambiguity in the query, based on the schema metadata.
15 . The apparatus as in claim 11 , wherein the schema metadata is for time series data in the database captured from the computer network.
16 . The apparatus as in claim 15 , wherein the schema metadata is indicative of at least one of: a number of records per time unit for the time series data or a maximum time window for the time series data.
17 . The apparatus as in claim 11 , wherein the query asks a status of a particular networking device in the computer network.
18 . The apparatus as in claim 11 , wherein the apparatus resolves any time ambiguity in the query using the schema metadata based in part on contextual data associated with a user that issued the query via the user interface.
19 . The apparatus as in claim 11 , wherein the process when executed is further configured to:
generate the answer to the query in part by translating the query into an application programming interface (API) call, after resolving any time ambiguity in the query using the schema metadata.
20 . A tangible, non-transitory, computer-readable medium storing program instructions that cause a device to execute a process comprising:
obtaining, by the device, a query from a user interface regarding a computer network;
retrieving, by the device and based on the query, schema metadata from a database;
resolving, by the device, any time ambiguity in the query using the schema metadata; and
providing, by the device, an answer to the query to the user interface, after resolving any time ambiguity in the query using the schema metadata.