Sensitive Data Classification via Feature Aggregation and ML
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for classifying sensitive data elements in files face challenges such as low accuracy, scalability issues, and high manual intervention, particularly when dealing with unstructured data and inconsistent file structures.
Innovation Solution
A system and method utilizing a processing subsystem with multiple modules, including a receiving module, conversion module, machine learning module, feature generation module, feature calculation module, comparison module, and classification module, to classify sensitive data elements by converting unstructured data into a machine-readable format, generating and aggregating sensitive data features, and using machine learning to predict true positive or false positive sensitive data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional data classification methods are used, then the process is simpler, but accuracy in classifying sensitive data elements deteriorates
Solution Approach 1:
The system segments the sensitive data classification process into multiple specialized modules: receiving module for data intake, conversion module for format transformation, feature generation module for extracting sensitive data characteristics, feature calculation module for computing feature metrics, comparison module for evaluating features against thresholds, and classification module for final categorization. This segmentation allows each module to focus on a specific aspect of classification, improving overall accuracy while managing complexity through modular design.
Solution Approach 2:
The system introduces feature generation and feature calculation modules as intermediaries between the raw sensitive data and the classification decision. These intermediary modules transform raw data into meaningful feature representations that capture the essential characteristics of sensitive data elements, enabling more accurate classification without requiring the classification module to directly process complex raw data.
2Productivity
If manual intervention is used for sensitive data classification, then accuracy may be maintained, but productivity deteriorates
Solution Approach 1:
The system implements self-service automation where the classification system autonomously performs data intake, conversion, feature extraction, feature calculation, comparison, and classification without requiring manual intervention at each step. The system serves itself by automatically processing sensitive data through the complete workflow, significantly improving productivity while maintaining accuracy through the sophisticated machine learning-based classification logic.
Solution Approach 2:
The system performs preliminary actions by pre-processing the sensitive data through conversion and feature generation before classification. The feature generation module extracts relevant characteristics in advance, and the feature calculation module computes necessary metrics beforehand, so that the final classification module receives pre-prepared data ready for rapid decision-making, improving both speed and accuracy.
3Reliability
If comprehensive sensitive data extraction is performed, then completeness improves, but processing time deteriorates
Solution Approach 1:
The system extracts only the essential features and characteristics relevant to sensitive data classification rather than processing all possible data attributes. The feature generation module selectively identifies and extracts key features that are most indicative of sensitive data types, while the feature calculation module focuses computation on these extracted features. This selective extraction maintains completeness of classification while reducing overall processing time by avoiding unnecessary analysis of irrelevant data elements.
Data Source
AI summary
A system and a method for classifying sensitive data elements in a file is provided. The method includes receiving and converting, the unstructured data file into a machine-readable format and generating, a plurality of sensitive data features. The plurality of sensitive data features represents single element of the sensitive data. The method includes generating, a plurality of adjacent elements corresponding to the single elements of the sensitive data and generating a plurality of feature categories. The method includes aggregating, the plurality of adjacent node features and the plurality of edge features. The method includes calculating and concatenating the plurality of aggregated adjacent nodes features and the plurality of aggregated edge features. The method includes comparing, the distance of the sensitive data from all of the adjacent sensitive data. The method includes classifying and predicting, the sensitive data to be a true positive or false positive sensitive data by using machine learning.