Generative LLM for Communication Network Data Transformation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing communication network operations lack efficient methods to transform numeric network operational data into a text-based format for training generative machine learning models, limiting the ability to generate insightful textual outputs in response to queries.

Innovation Solution

A processing system that obtains network operational data, transforms it into a text-based format, and trains a generative machine learning model using this transformed data, enabling the generation of textual outputs in response to queries.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If network operational data is maintained in database tables (structured format), then data storage and retrieval efficiency is improved, but the ability to train generative machine learning models is limited

Engineering Contradiction:

Improvedata retrieval efficiencyVSAvoidmodel training capability

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent segments the data representation into two formats: structured database tables for storage and retrieval operations, and text-based representations for machine learning model training. This segmentation allows each format to optimize for its specific purpose without compromising the other.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary transformation layer that converts network operational data from structured database format to text-based format. This intermediary enables the bridge between database systems and generative machine learning models, allowing both structured data storage and unstructured data processing to coexist.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Adaptability or versatility

If network operational data is transformed into text-based format for generative model training, then the ability to generate insightful textual outputs is improved, but the complexity of data processing increases

Engineering Contradiction:

Improvetextual output generation capabilityVSAvoiddata processing complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements preliminary action by pre-transforming network operational data into text-based format before it is needed for model training. This pre-processing step creates ready-to-use text representations that simplify the overall data processing workflow and reduce real-time computational complexity.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent creates a textual copy or representation of the network operational data that can be independently processed by generative models. This copy allows the original structured data to remain intact in databases while providing a simplified version for AI processing.

Inventive Principle:

Principle #26Copying

3Ease of manufacture

If existing database tables are used for network operational data, then data storage is simplified, but the ability to provide timely and accurate insights is limited

Engineering Contradiction:

Improvedata storage simplicityVSAvoidinsight generation speed

Core Design Contradiction:

Ease of manufactureVSProductivity

Solution Approach 1:

The patent replaces traditional mechanical database querying mechanisms with generative machine learning models that can directly process text-based representations of network data. This substitution enables faster and more flexible insight generation without requiring complex SQL queries or data aggregation operations.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS20250200086A1Communication network management using generative large language model

Publication Date: 2025.06.19 AT&T INTELLECTUAL PROPERTY I L P

AI summary

A processing system including at least one processor may obtain network operational data of a communication network, transform the network operational data into a text-based format, and train a generative machine learning model implemented by the processing system using the network operational data in the text-based format. The processing system may then receive a query pertaining to the network operational data, apply the query to the generative machine learning model implemented by the processing system to generate a textual output in response to the query, and present the textual output that is generated in response to the query.