Communication network data management and visualization using generative large language model-based query statement generation

1 . A method comprising:

obtaining, by a processing system including at least one processor, a natural language request associated with a database system;

generating, by the processing system, a prompt based upon the natural language request in accordance with a prompt mapping function, wherein the prompt mapping function comprises a term mapping function that matches terms in natural language requests to data fields of data tables of the database system, wherein the term mapping function comprises: a thesaurus or an ontology, and wherein the thesaurus or the ontology comprises a feature graph that includes: nodes and edges, each of the nodes representing a respective one of the data fields of the data tables of the database system and each of the edges representing a respective pair of the data fields of the data tables of the database system being related, and wherein the generating of the prompt includes: generating an output of the prompt mapping function in response to the natural language request as an input, truncating, from the output, a portion of the output that relates to visualization to generate a truncated output, and generating the prompt by inserting the truncated output into a prompt template;

applying, by the processing system, the prompt as an input to a generative model implemented by the processing system to generate a structured query, wherein the applying includes embedding a sub-graph of the thesaurus or the ontology with the prompt as the input to the generative model;

applying, by the processing system, the structured query to the database system to obtain a query result; and

presenting, by the processing system, at least one visualization of the query result.

2 . The method of claim 1 , further comprising:

generating the at least one visualization of the query result.

3 . The method of claim 2 , wherein the at least one visualization of the query comprises at least one result table.

4 . The method of claim 2 , wherein the at least one visualization comprises at least one of:

a chart; or

a graph.

5 . The method of claim 2 , wherein the natural language request includes a visualization request.

6 . The method of claim 5 , wherein the at least one visualization of the query result is generated in response to the visualization request.

7 . The method of claim 1 , wherein the database system comprises data tables of communication network operational data.

8 . The method of claim 7 , wherein the processing system is deployed in a communication network.

9 . The method of claim 1 , wherein the generative model comprises a large language model-based machine learning model.

10 . The method of claim 1 , wherein the generative model comprises a generative pre-trained transformer model.

11 . The method of claim 1 , wherein the applying of the prompt as the input to the generative model to generate the structured query further comprises:

generating supplemental prompt content from one or more data tables of the database system; and

applying the supplemental prompt content as an additional input to the generative model.

12 . The method of claim 1 , wherein the data fields comprise at least one of:

table columns; or

fields of data elements within a table column.

13 . The method of claim 12 , wherein the data elements comprise java script object notation elements.

14 . The method of claim 1 , wherein the term mapping function is updated via machine learning.

15 . The method of claim 1 , wherein the database system comprises a structured query language-based system.

16 . A non-transitory computer-readable medium storing instructions which, when executed by a processing system including at least one processor, cause the processing system to perform operations, the operations comprising:

obtaining a natural language request associated with a database system;

generating a prompt based upon the natural language request in accordance with a prompt mapping function, wherein the prompt mapping function comprises a term mapping function that matches terms in natural language requests to data fields of data tables of the database system, wherein the term mapping function comprises: a thesaurus or an ontology, and wherein the thesaurus or the ontology comprises a feature graph that includes: nodes and edges, each of the nodes representing a respective one of the data fields of the data tables of the database system and each of the edges representing a respective pair of the data fields of the data tables of the database system being related, and wherein the generating of the prompt includes: generating an output of the prompt mapping function in response to the natural language request as an input, truncating, from the output, a portion of the output that relates to visualization to generate a truncated output, and generating the prompt by inserting the truncated output into a prompt template;

applying the prompt as an input to a generative model implemented by the processing system to generate a structured query, wherein the applying includes embedding a sub-graph of the thesaurus or the ontology with the prompt as the input to the generative model;

applying the structured query to the database system to obtain a query result; and

presenting at least one visualization of the query result.

17 . An apparatus comprising:

a processing system including at least one processor; and

a computer-readable medium storing instructions which, when executed by the processing system, cause the processing system to perform operations, the operations comprising:

obtaining a natural language request associated with a database system;

generating a prompt based upon the natural language request in accordance with a prompt mapping function, wherein the prompt mapping function comprises a term mapping function that matches terms in natural language requests to data fields of data tables of the database system, wherein the term mapping function comprises: a thesaurus or an ontology, and wherein the thesaurus or the ontology comprises a feature graph that includes: nodes and edges, each of the nodes representing a respective one of the data fields of the data tables of the database system and each of the edges representing a respective pair of the data fields of the data tables of the database system being related, and wherein the generating of the prompt includes: generating an output of the prompt mapping function in response to the natural language request as an input, truncating, from the output, a portion of the output that relates to visualization to generate a truncated output, and generating the prompt by inserting the truncated output into a prompt template;

applying the prompt as an input to a generative model implemented by the processing system to generate a structured query;

applying the structured query to the database system to obtain a query result, wherein the applying includes embedding a sub-graph of the thesaurus or the ontology with the prompt as the input to the generative model; and

presenting at least one visualization of the query result.

18 . The apparatus of claim 17 , wherein the operations further comprise:

generating the at least one visualization of the query result.

19 . The apparatus of claim 18 , wherein the at least one visualization of the query comprises at least one result table.

20 . The apparatus of claim 18 , wherein the at least one visualization comprises at least one of:

a chart; or

a graph.