Software Security Integrity via NLP Requirement Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing software development processes struggle to integrate security requirements effectively, leading to a philosophical and practical separation between business and technical user descriptions and security concerns, resulting in high false-positive and false-negative results, and a lack of scalability in expert resources.
Innovation Solution
Utilizing machine learning models, specifically Natural Language Processing (NLP) and deep learning techniques, to classify functional requirements and automatically generate security acceptance criteria and features, integrating them into the software development lifecycle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If manual security review processes are used, then security concerns can be addressed, but the process lacks scalability and expertise is limited
Solution Approach 1:
The patent replaces manual security review processes with an automated machine learning-based system. The system uses NLP to classify functional requirements and deep learning models to generate security acceptance criteria, eliminating the need for manual expert review and enabling scalable security integration across multiple projects simultaneously.
Solution Approach 2:
The system performs self-service by automatically analyzing functional requirements, classifying security concerns, and generating acceptance criteria without human intervention. The machine learning models continuously improve by learning from project data, enabling the system to autonomously maintain high security standards across evolving software projects.
2Measurement precision
If security requirements are integrated into development processes, then security accuracy improves, but false-positive and false-negative results increase due to separation between business and technical descriptions
Solution Approach 1:
The patent merges business functional requirements with security concerns by using NLP to process both types of information together. The system classifies functional requirements and maps them to security implications, ensuring that security assessment considers both business context and technical details without information loss.
Solution Approach 2:
The machine learning classification model acts as an intermediary between business functional requirements and security concerns. It processes functional requirement descriptions, identifies security implications, and generates appropriate acceptance criteria, bridging the gap between business and security domains while maintaining accuracy.
3Reliability
If expert resources are increased to improve security coverage, then security quality improves, but resource scalability is limited
Solution Approach 1:
The patent substitutes human expert resources with an automated machine learning system that provides consistent security coverage. The deep learning models analyze functional requirements and generate security acceptance criteria with reliability comparable to expert review, eliminating the need to scale human resources while maintaining or improving security quality.
Data Source
AI summary
A method may include querying, using a processing unit, a project data store with a project identifier; in response to the querying, receiving a functional requirement of a project data structure stored as associated with the project identifier; inputting, using the processing unit, the functional requirement into a trained machine learning model, the machine learning model configured with output nodes corresponding to a set of requirement classifications; after the inputting, accessing, using the processing unit, output values of the output nodes; and adding, using the processing unit, a requirement classification of the set of requirement classifications based on the output values to the project data structure.