Machine Learning Security Scan Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing software scanning technologies generate substantial 'noise' in scan results due to findings that are not relevant to the application under test, making it difficult to identify and filter out irrelevant security vulnerabilities.
Innovation Solution
A machine learning process is used to identify and filter security requirements by receiving input on security characteristics, building a list of security requirements, and applying machine learning to add or remove requirements, followed by a security vulnerability scan using the updated list.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If comprehensive security scanning is performed to identify all potential vulnerabilities, then security coverage is improved, but scan noise increases making it difficult to identify relevant vulnerabilities
Solution Approach 1:
The patent segments the comprehensive scan results into relevant and irrelevant portions by analyzing security requirements specific to the application's technology stack, framework, and configuration. This segmentation allows the system to filter out noise while maintaining coverage of actual security risks.
Solution Approach 2:
The patent introduces an intermediary analysis layer that sits between the comprehensive security scan and the final results presented to users. This intermediary layer uses machine learning models trained on security requirements to filter and prioritize findings, reducing noise while preserving relevant vulnerability information.
2Productivity
If machine learning is used to filter scan results to reduce noise, then scan efficiency is improved, but complexity of the scanning system increases
Solution Approach 1:
The patent implements self-service through automated machine learning models that automatically learn and adapt to the specific application's security requirements without requiring manual configuration or expert intervention. The system serves itself by autonomously filtering scan results based on learned patterns from the application's technology stack and configuration.
Solution Approach 2:
The patent changes the parameters of the scanning system by introducing machine learning models that dynamically adjust filtering criteria based on the application's specific characteristics. Instead of using fixed, static filtering rules, the system adapts its parameters (thresholds, priority levels, relevance criteria) based on learned patterns from the application's security requirements.
Data Source
AI summary
An input regarding security characteristics of a project is received. For example, a security characteristic of a project may be insecure storage of data related to confidentiality. The project is scanned for one or more security requirements based on the received security characteristics. A list of security requirements is built for the project based on the received first input. A machine learning process is used to identify addition of one or more security requirements and/or removal of one or more security requirements from the list of security requirements. A first security vulnerability scan is run using the list of security requirements with the one or more additional security requirements and/or the removed one or more security requirements. Results for the first security vulnerability scan are generated and displayed to a user.