Granted Patent B2

US 12,541,653 · App. 17/513,419 · Granted Feb 3, 2026

Inventors: Tomasz Ploskon (Cracow, PL); Filis Omer (Constanta, RO); Costel Moraru (Egmating, DE); Laurentiu Gabriel Ghergu (Bucharest, RO); Erik Rueger (Ockenheim, DE)

Assignee: International Business Machines Corporation

G06F40/40G06Q10/06375

Abstract

Claims (70)

1 . A system comprising:

a processor set;

one or more computer-readable storage media; and

program instructions stored on the one or more computer-readable storage media to cause the processor set to perform operations comprising:

analyzing a communication associated with a simulation program;

determining that the simulation program is running;

accessing network traffic between a first natural language processing provider and the simulation program using a network connection and an application program interface of the first natural language processing provider;

capturing, through network traffic analysis at an application layer, at least one request/response pair in the communication from the network traffic;

extracting metadata associated with the request/response pair from the network traffic wherein the metadata includes an identified intent, intent confidence level, and entities;

storing the metadata and the at least one request/response pair in a persistent storage;

training at least one second registered natural language processing provider with the metadata and the request/response pair to replicate behavior of the first natural language processing provider; and

migrating requests from the first natural language processing provider to the at least one second natural language processing provider after determining the at least one second natural language processing provider responds within a confidence threshold of the first natural language processing provider.

2 . The system of claim 1 , wherein the processor is further configured to perform operations comprising:

determining if an amount of the at least one request/response pair is above an NLP threshold.

3 . The system of claim 1 , wherein training the at least one registered natural language processing provider is automatic.

4 . The system of claim 1 , wherein the processor is further configured to perform operations comprising:

analyzing an interaction of an operator to capture the at least one request/response pair; and

simulating the interaction.

5 . The system of claim 1 , wherein the communication is analyzed on an application layer, and wherein the system further comprises:

simulating remote cloud natural language processing services of the at least one natural language processing provider in an on-premise environment; and

predicting a future action of the remote cloud natural language processing services.

6 . The system of claim 5 , wherein the processor is further configured to perform operations comprising:

generating one or more recommendations associated with the remote natural language processing services, wherein the one or more recommendations are based on the future action; and

providing the one or more recommendations to the at least one natural language processing provider.

7 . The system of claim 5 , wherein the processor is further configured to perform operations comprising:

reducing a total number of requests to the remote cloud natural language processing services of the at least one natural language processing services.

8 . A method comprising:

analyzing, by a processor, a communication associated with a simulation program;

determining that the simulation program is running;

accessing network traffic between a first natural language processing provider and the simulation program using a network connection and an application program interface of the first natural language processing provider;

capturing, through network traffic analysis at an application layer, at least one request/response pair in the communication from the network traffic;

extracting metadata associated with the request/response pair from the network traffic wherein the metadata includes an identified intent, intent confidence level, and entities;

storing the metadata and the at least one request/response pair in persistent storage;

training at least one second registered natural language processing provider with the metadata and the request/response pair to replicate behavior of the first natural language processing provider; and

migrating requests from the first natural language processing provider to the at least one second natural language processing provider after determining the at least one second natural language processing provider responds within a confidence threshold of the first natural language processing provider.

9 . The computer-implemented method of claim 8 , further comprising:

determining if an amount of the at least one request/response pair is above an NLP threshold.

10 . The computer-implemented method of claim 8 , wherein training the at least one registered natural language processing provider is automatic.

11 . The computer-implemented method of claim 8 , further comprising:

analyzing an interaction of an operator to capture the at least one request/response pair; and

simulating the interaction.

12 . The computer-implemented method of claim 8 , wherein the communication is analyzed on an application layer, and wherein the method further comprises:

simulating remote cloud natural language processing services of the at least one natural language processing provider in an on-premise environment; and

predicting a future action of the remote cloud natural language processing services.

13 . The computer-implemented method of claim 12 , further comprising:

generating one or more recommendations associated with the remote natural language processing services, wherein the one or more recommendations are based on the future action; and

providing the one or more recommendations to the at least one natural language processing provider.

14 . The computer-implemented method of claim 12 , further comprising:

reducing a total number of requests to the remote cloud natural language processing services of the at least one natural language processing services.

15 . A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform operations comprising:

analyzing a communication associated with a simulation program;

determining that the simulation program is running;

accessing network traffic between a first natural language processing provider and the simulation program using a network connection and an application program interface of the first natural language processing provider;

capturing, through network traffic analysis at an application layer, at least one request/response pair in the communication from the network traffic;

extracting metadata associated with the request/response pair from the network traffic wherein the metadata includes an identified intent, intent confidence level, and entities;

storing the metadata and the at least one request/response pair in persistent storage;

training at least one second registered natural language processing provider with the metadata and the request/response pair to replicate behavior of the first natural language processing provider; and

migrating requests from the first natural language processing provider to the second natural language processing provider after determining the second natural language processing provider responds within a confidence threshold of the first natural language processing provider.

16 . The computer program product of claim 15 , wherein the processor is further configured to perform operations comprising:

determining if an amount of the at least one request/response pair is above an NLP threshold.

17 . The computer program product of claim 15 , wherein training the at least one registered natural language processing provider is automatic.

18 . The computer program product of claim 15 , wherein the processor is further configured to perform operations comprising:

analyzing an interaction of an operator to capture the at least one request/response pair; and

simulating the interaction.

19 . The computer program product of claim 15 , wherein the communication is analyzed on an application layer, and wherein the processor is further configured to perform operations comprising:

simulating remote cloud natural language processing services of the at least one natural language processing provider in an on-premise environment; and

predicting a future action of the remote cloud natural language processing services.

20 . The computer program product of claim 19 , wherein the processor is further configured to perform operations comprising:

generating one or more recommendations associated with the remote natural language processing services, wherein the one or more recommendations are based on the future action; and

providing the one or more recommendations to the at least one natural language processing provider.

Assignments (1)

Continuity (1)

References Cited (28)