Cyber Risk Classification Using Probabilistic Neural Network
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods lack objective and quick risk assessment tools for determining whether textual documents transiting computer networks contain malicious code, relying heavily on subjective evaluations and lacking automated, quantitative decision-making processes.
Innovation Solution
A system utilizing a Probabilistic Neural Network (PNN) to classify textual documents based on their cyber risk, integrating with publicly available threat databases like CVE, and employing a bag-of-words approach with stemming and stop words elimination for effective risk evaluation and routing decisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If subjective expert evaluation is used for risk assessment, then flexibility in decision-making is maintained, but objectivity and speed of decision are compromised
Solution Approach 1:
The patent replaces subjective human expert evaluation with an automated Probabilistic Neural Network (PNN) system that objectively assesses cyber risk. The PNN processes document features through neural network layers to produce quantitative risk scores, eliminating human subjectivity while maintaining decision flexibility through learned patterns from training data.
Solution Approach 2:
The system transforms qualitative risk assessment into quantitative parameters by extracting numerical features from documents (word frequency, document length, metadata) and processing them through the PNN to generate standardized risk scores. This parameter transformation enables objective comparison and automated decision-making.
2Measurement precision
If comprehensive vulnerability analysis is performed, then risk assessment accuracy is improved, but processing time and system response speed deteriorate
Solution Approach 1:
The PNN model is pre-trained on extensive vulnerability data and document patterns before actual risk assessment occurs. This preliminary training allows the system to quickly evaluate new documents without performing exhaustive analysis from scratch, achieving both accuracy and speed by leveraging pre-learned patterns.
Solution Approach 2:
The system extracts only the most relevant features from documents for risk assessment (such as word frequency, document metadata, and key vocabulary patterns) rather than analyzing every aspect comprehensively. This feature selection enables fast processing while maintaining assessment accuracy through focus on critical risk indicators.
3Productivity
If automated routing decisions are implemented, then productivity and speed are improved, but reliability and accountability may be reduced
Solution Approach 1:
The system incorporates feedback mechanisms where routing decisions can be reviewed and adjusted based on actual outcomes and new vulnerability information. The PNN continuously learns from new data and can have its parameters adjusted, ensuring that automated decisions remain reliable and adaptable to evolving cyber threats while maintaining high processing speed.
Data Source
AI summary
Disclosed herein are computer-implement systems and methods for determining whether an electronic document transiting a computer network is safe for viewing and does not contain malicious code, and automatically routing or deleting such electronic document based on such determination, thus providing quick risk decisions relating to the electronic document using an objective, quantitative method. A search algorithm is employed to detect the presence of words or phrases in the subject electronic document that are possible threats to the computer network. Threat updates may be provided through publicly available threat database. A weight is added to the generated score that accounts for the time that the particular vulnerability has been in the subject threat database. Finally, a PNN is used to quickly and objectively classify the subject electronic document for acceptance, quarantine, or denial.