ML-Based Regulatory Requirement Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity and volume of regulatory content make it challenging for companies to identify and comply with specific requirements, as existing methods rely on manual review by trained staff and provide only high-level descriptions without detailed action items.
Innovation Solution
A computer-implemented method and system that uses machine learning models to classify, split, and merge regulatory content, extracting specific requirements by analyzing citations and their relationships, and presenting them in a structured format for compliance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual review by trained staff is used to identify regulatory requirements, then high-level descriptions can be provided, but detailed action items cannot be extracted and the process is highly time-consuming
Solution Approach 1:
The patent replaces the mechanical manual review process with an automated machine learning system. The classifier model automatically processes regulatory text to extract requirements, classifications, and action items without human intervention, thereby eliminating the time-consuming manual review while maintaining or improving extraction precision through consistent application of classification rules.
Solution Approach 2:
The system enables self-service extraction of regulatory requirements through the automated machine learning pipeline. The model independently performs classification, requirement extraction, and action item identification without requiring trained staff to manually analyze each regulatory document, allowing the system to serve itself in processing and extracting compliance information.
2Adaptability or versatility
If the volume of regulatory content increases to cover more regulations, then compliance coverage is improved, but the complexity and difficulty of identifying specific requirements increases
Solution Approach 1:
The patent applies segmentation by breaking down the complex regulatory content into discrete, manageable units. The machine learning model processes regulatory text by identifying individual requirements, classifications, and action items as separate elements. This segmentation allows the system to handle large volumes of diverse regulations by treating each requirement as an independent extractable unit, thereby improving compliance coverage without proportionally increasing system complexity.
3Loss of information
If existing manual methods are used to review regulatory content, then high-level descriptions can be obtained, but specific actionable requirements cannot be systematically extracted
Solution Approach 1:
The patent replaces manual review mechanics with an automated machine learning system that systematically extracts complete requirement information. The classifier model processes the full regulatory text to identify and extract all actionable requirements, classifications, and details, thereby eliminating information loss while implementing high-level automation in the extraction process.
Data Source
AI summary
Described herein are systems and methods for extracting requirements from regulatory content data. The method including: receiving the regulatory content data; classifying an associated type for each citation in the regulatory content data using a trained classifier machine learning model, the classifier machine learning model trained using regulatory content data including expert labelled annotations; splitting citations in the regulatory content data, including determining whether each citation includes more than one requirement; merging one or more citations in the regulatory content data, including identifying child-parent relationships for the citations and merging citations based on conjunctive language; and outputting the citations and their associated type. In a particular case, the types of citations for classification include one of a requirement (REQ), an optional or site-specific requirement (OSR), a description (DSC), and part of another requirement.