Machine Learning System for Structuring Unstructured Technical Reports

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Converting unstructured technical reports to structured formats is labor-intensive, inefficient, and prone to errors due to the variability in terminology and grammatical structure, making it difficult to extract and leverage information effectively.

Innovation Solution

A machine learning-based system that identifies technical entities and builds logical relationships between them, using natural language processing and domain knowledge to generate structured reports, with a graphical user interface for user input and refinement to adapt the pattern matching algorithm.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual data entry is used to convert unstructured technical reports to structured formats, then information can be extracted and organized, but the process becomes labor-intensive, time-consuming, and prone to errors

Engineering Contradiction:

Improvedata extraction accuracyVSAvoidconversion time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical data entry with an automated natural language processing system that uses machine learning models to extract technical entities and relationships from unstructured text, eliminating human labor while maintaining high accuracy through trained algorithms

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service conversion where the NLP model automatically processes technical reports without human intervention, using trained patterns to identify entities, relationships, and structured information directly from unstructured input documents

Inventive Principle:

Principle #25Self-service

2Productivity

If manual conversion methods are used, then some level of data extraction is achieved, but consistency and reliability deteriorate due to variability in terminology and grammatical structure

Engineering Contradiction:

Improvedata extraction efficiencyVSAvoidextraction consistency

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent transforms unstructured text parameters into structured data parameters through NLP processing, changing the state of information from free-form text to organized entities with standardized attributes, enabling consistent extraction across varied input formats

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system replaces inconsistent manual extraction with automated NLP processing that applies consistent rules and trained models to identify technical entities and relationships, ensuring uniform extraction quality regardless of input variability

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Adaptability or versatility

If information is stored in unstructured formats, then flexibility in report creation is maintained, but information accessibility and usability deteriorate

Engineering Contradiction:

Improvereport format flexibilityVSAvoidinformation accessibility

Core Design Contradiction:

Adaptability or versatilityVSEase of operation

Solution Approach 1:

The patent segments unstructured text into distinct technical entities and relationships, organizing information into discrete structured components that can be independently accessed, queried, and utilized while preserving the flexibility of the original report format

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The NLP processing system acts as an intermediary that translates between unstructured flexible formats and structured accessible formats, enabling users to query and analyze data systematically while the underlying reports maintain their original flexible presentation

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11790170B2Converting unstructured technical reports to structured technical reports using machine learning

Publication Date: 2023.10.17 CHEVRON USA INC

AI summary

A computer-implemented, machine learning-based method of converting an unstructured technical report into a structured technical report includes obtaining an unstructured technical report, tokenizing the unstructured technical report into an n-gram array, identifying and filtering non-interesting n-grams from the first n-gram array based on common language usage of the non-interesting n-grams and a determination that the non-interesting n-grams do not appear on a confirmed technical entity database, generating and displaying a technical entity candidate list from the filtered n-gram array, displaying, obtaining, from a pattern matching model and/or a graphical user interface, an indication that a technical entity candidate is a technical entity of interest, appending the technical entity of interest to the confirmed technical entity database, generating and displaying a structured technical report with the confirmed technical entities and corresponding technical entity value parameters, and iterating the process to refine the pattern matching model.