Intelligent auto-prompt engine for network management
A unified prompt-based network management system that involves an intelligent auto-prompt engine generating contextualized prompts for an artificial intelligence model. The artificial intelligence model generates instructions and/or solutions and adapts to different application scenarios based on an enterprise network knowledge and reverse inference(s). Specifically, methods are provided that involve obtaining input data related to a configuration or an operation of one or more assets in an enterprise network and generating a contextualized prompt based on the input data, network knowledge information of the enterprise network, and at least one reverse inference generated using an artificial intelligence model. The methods further involve providing the contextualized prompt to the artificial intelligence model for generating a tailored response to the input data, wherein the tailored response includes a set of actionable tasks to be performed with respect to the one or more assets of the enterprise network.
1 . A computer-implemented method comprising:
obtaining input data related to a configuration or an operation of one or more assets in an enterprise network;
generating a contextualized prompt based on the input data, network knowledge information of the enterprise network, and at least one reverse inference generated using an artificial intelligence model;
providing the contextualized prompt to the artificial intelligence model for generating a tailored response to the input data, wherein the tailored response includes a set of actionable tasks to be performed with respect to the one or more assets of the enterprise network; and
changing, by a computing device, the configuration of at least one affected network device of the one or more assets in the enterprise network by performing the set of actionable tasks.
2 . The computer-implemented method of claim 1 , wherein the input data is a user input and further comprising:
obtaining the network knowledge information, which includes a topology of the enterprise network and data related to a plurality of network features of a plurality of assets in the enterprise network; and
generating at least one enterprise network embedding based on the network knowledge information, wherein the contextualized prompt is generated by processing the input data based on the at least one enterprise network embedding.
3 . The computer-implemented method of claim 1 , wherein the input data is a user input and the network knowledge information include a network knowledge graph.
4 . The computer-implemented method of claim 3 , further comprising:
extracting a set of nodes from a plurality of nodes of the network knowledge graph based on the input data; and
generating a network knowledge vector based on the set of nodes, wherein the contextualized prompt is generated by processing the input data based on the network knowledge vector.
5 . The computer-implemented method of claim 1 , wherein the input data includes at least one query or instruction from a user, and generating the contextualized prompt further includes:
generating at least one user embedding based on a persona of the user which includes one or more of a user profile and a user role within the enterprise network; and
generating the contextualized prompt by processing the input data based on the at least one user embedding.
6 . The computer-implemented method of claim 1 , wherein changing the configuration of the at least one affected network device includes:
establishing, by the computing device, a connection with each of the at least one affected network device using an application programming interface; and
reconfiguring, by the computing device, a hardware or a firmware on a respective network device.
7 . The computer-implemented method of claim 1 , further comprising training the artificial intelligence model during an inference phase to generate the at least one reverse inference.
8 . A computer-implemented method comprising:
obtaining input data related to a configuration or an operation of one or more assets in an enterprise network;
generating a contextualized prompt based on the input data, network knowledge information of the enterprise network, and at least one reverse inference generated using an artificial intelligence model;
providing the contextualized prompt to the artificial intelligence model for generating a tailored response to the input data, wherein the tailored response includes a set of actionable tasks to be performed with respect to the one or more assets of the enterprise network; and
training the artificial intelligence model during an inference phase by:
obtaining a prompt including context information about the enterprise network and user input,
generating a solution including at least one actionable task by processing the prompt using the artificial intelligence model,
generating a new prompt including the context information and the solution, and
generating the at least one reverse inference by processing the new prompt using the artificial intelligence model.
9 . The computer-implemented method of claim 8 , further comprising:
training the artificial intelligence model based on a plurality of network domain specific applications, wherein generating the contextualized prompt is further based on a selected one of the plurality of network domain specific applications.
10 . The computer-implemented method of claim 9 , wherein the plurality of network domain specific applications include a network connectivity related service, a network security related service, and a network management service and wherein the artificial intelligence model is a large language machine model.
11 . The computer-implemented method of claim 8 , wherein the input data includes at least one query or instruction from a user, and wherein generating the contextualized prompt includes:
generating at least one user embedding based on a persona of the user which includes one or more of a user profile and a user role within the enterprise network; and
generating the contextualized prompt by processing the input data based on the at least one user embedding.
12 . An apparatus comprising:
a memory;
a network interface configured to enable network communications; and
a processor, wherein the processor is configured to perform a method comprising:
obtaining input data related to a configuration or an operation of one or more assets in an enterprise network;
generating a contextualized prompt based on the input data, network knowledge information of the enterprise network, and at least one reverse inference generated using an artificial intelligence model, wherein the contextualized prompt is generated by:
generating input embedding based on the input data,
generating at least one network embedding based on the network knowledge information in a network knowledge graph and the input data, and
generating, by an artificial intelligence prompt engine, the contextualized prompt based on the input embedding, the at least one network embedding, and the at least one reverse inference;
generating, by the artificial intelligence model, a tailored response to the input data based on the contextualized prompt, wherein the tailored response includes a set of actionable tasks to be performed with respect to the one or more assets of the enterprise network; and
providing the tailored response for configuring the one or more assets in the enterprise network.
13 . The apparatus of claim 12 , wherein the input data is a user input and the processor is further configured to perform:
obtaining the network knowledge information, which includes a topology of the enterprise network and data related to a plurality of network features of a plurality of assets in the enterprise network; and
generating the at least one network embedding based on the network knowledge information.
14 . The apparatus of claim 12 , wherein the input data is a user input and the network knowledge information include the network knowledge graph.
15 . The apparatus of claim 14 , wherein the processor is further configured to perform:
extracting a set of nodes from a plurality of nodes of the network knowledge graph based on the input data; and
generating a network knowledge vector based on the set of nodes, wherein the contextualized prompt is generated by processing the input data based on the network knowledge vector.
16 . The apparatus of claim 12 , wherein the input data includes at least one query or instruction from a user, and wherein the processor is configured to generate the contextualized prompt further by:
generating at least one user embedding based on a persona of the user which includes one or more of a user profile and a user role within the enterprise network; and
generating the contextualized prompt by processing the input data based on the at least one user embedding.
17 . The apparatus of claim 12 , wherein the processor is further configured to perform:
changing, by a computing device, the configuration of one or more affected network devices in the enterprise network by performing the set of actionable tasks.
18 . The apparatus of claim 12 , further comprising:
generating, by the artificial intelligence model, the at least one reverse inference based on a known set of actionable tasks.
19 . The apparatus of claim 12 , wherein the at least one reverse inference is generated by:
obtaining a prompt including context information about the enterprise network and user input;
generating a solution including at least one actionable task by processing the prompt using the artificial intelligence model;
generating a new prompt including the context information and the solution; and
generating the at least one reverse inference by processing the new prompt using the artificial intelligence model.
20 . The apparatus of claim 12 , wherein generating the contextualized prompt further includes:
adopting, by the artificial intelligence prompt engine, a prompt to a specific network domain based on a user persona, to generate the contextualized prompt.