Data Value Classifier for Dynamic Network Asset Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data loss prevention (DLP) systems are unable to effectively identify valuable data and assets on a network, as they rely on static patterns and fail to recognize dynamically changing data and asset risks.
Innovation Solution
The implementation of a data value classifier using machine learning models, such as convolutional neural networks, to classify data as valuable or non-valuable, and an asset risk classifier to determine if an asset is a target based on the classified data, along with the ability to update the models with user feedback and classification rules.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional DLP systems use a fixed set of patterns to classify data, then the system structure is simple and easy to implement, but the system cannot recognize dynamically changing data or assets that should be scrutinized and protected
Solution Approach 1:
The patent applies the Dynamics principle by transitioning from static pattern-matching DLP systems to dynamic machine learning models that continuously learn and adapt to new data types and threats. The system uses trained classifiers that can dynamically identify valuable data patterns without requiring manual pattern updates, resolving the contradiction between adaptability and complexity by automating the learning process.
Solution Approach 2:
The system implements Self-service through automated machine learning models that independently classify and identify valuable data without requiring continuous manual configuration. The models self-adjust to new data patterns through training processes, reducing the need for manual system intervention while maintaining high adaptability to changing data landscapes.
2Measurement precision
If machine learning models are implemented to accurately classify valuable data, then the classification accuracy is improved, but the computational resources and processing time are increased
Solution Approach 1:
The patent applies Preliminary action by pre-training machine learning models offline with extensive datasets before deployment. This preliminary training phase allows the models to achieve high classification accuracy in advance, so that during actual operation, the computational resources required are significantly reduced while maintaining precise classification of valuable data.
3Reliability
If comprehensive data analysis is performed to identify all valuable assets, then the security coverage is improved, but the processing time and system load are increased
Solution Approach 1:
The system applies Local quality by prioritizing classification and analysis of data based on its identified value and sensitivity. Rather than uniformly processing all data with the same computational resources, the system focuses intensive analysis on data classified as valuable or sensitive, while applying lighter processing to non-critical data, thus achieving comprehensive security coverage with optimized processing time.
Data Source
AI summary
Methods and systems for identifying targets on a network. The disclosed methods involve classifying data as valuable or non-valuable, and then classifying an asset associated with the retrieved data as a target or a non-target based in part on the classification of the data.