Database Metadata Pattern Matching for Sensitive Data Discovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing database systems face challenges in efficiently identifying and securing sensitive data due to the resource-intensive nature of reading and analyzing structured data, which can introduce security vulnerabilities and complicate data restoration processes.
Innovation Solution
A system utilizes database metadata to identify sensitive data types without accessing the data itself, using pattern matching and metadata rules to determine the locations of sensitive data within a database, enhancing visibility and accuracy of data restoration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system reads and analyzes structured data to identify sensitive data, then identification accuracy is improved, but resource consumption increases and security vulnerabilities are introduced
Solution Approach 1:
The patent extracts and analyzes database metadata (table names, column names, data types, constraints) separately from the actual structured data. This allows the system to identify sensitive data locations using minimal resources while maintaining identification accuracy through pattern matching on metadata attributes rather than scanning entire data sets.
Solution Approach 2:
The patent introduces metadata as an intermediary layer between the identification system and the structured data. By analyzing metadata patterns (e.g., column names containing 'ssn', 'credit_card', data types like 'numeric' with specific formats), the system can identify sensitive data without directly accessing or reading the actual data content, thus reducing resource consumption and security risks.
2Measurement precision
If the system reads and analyzes structured data to identify sensitive data, then identification accuracy is improved, but security vulnerabilities are introduced
Solution Approach 1:
The system extracts identification information from metadata rather than accessing structured data directly. This extraction approach allows sensitive data identification while minimizing exposure to security vulnerabilities associated with data access, as the system only handles metadata which contains structural information but not actual sensitive values.
Solution Approach 2:
Metadata serves as a secure intermediary that enables identification without direct data access. The pattern matching engine analyzes metadata patterns (column names, data types, constraints) to identify sensitive data locations without requiring privileged access to the actual data, thereby reducing security vulnerabilities while maintaining identification accuracy.
3Measurement precision
If the system accesses structured data for analysis, then data restoration accuracy is improved, but the complexity of data restoration processes increases
Solution Approach 1:
The system performs preliminary analysis of metadata to identify sensitive data locations, data types, and patterns before restoration operations. This preliminary metadata-based classification creates a roadmap for restoration, enabling accurate restoration of sensitive data while simplifying the overall process by avoiding the need to analyze actual data content during restoration.
Solution Approach 2:
The patent replaces complex mechanical data analysis operations with metadata-based pattern matching. Instead of reading and analyzing structured data to understand its content and context, the system uses metadata patterns (column names, data types, constraints) to automatically identify sensitive data and guide restoration, significantly reducing process complexity while maintaining accuracy.
4Loss of information
If the system performs comprehensive data analysis, then identification completeness is improved, but processing time increases
Solution Approach 1:
The patent segments the identification process into metadata analysis and pattern matching operations. By dividing the database structure into analyzable metadata components (tables, columns, data types, constraints) and applying pattern matching rules to each segment, the system achieves comprehensive identification of sensitive data across the entire database without the time cost of analyzing every data record.
Solution Approach 2:
The system replaces time-consuming mechanical data scanning with efficient metadata-based pattern matching. By substituting direct data analysis with metadata pattern recognition (e.g., identifying columns with names matching sensitive data patterns, or data types consistent with sensitive information), the system achieves complete identification while dramatically reducing processing time.
Data Source
AI summary
Techniques for database management are described. A database management system may transmit a request for a data management system of a database to provide a set of metadata attributes for structured data within the database, and may receive a set of metadata attributes for the structured data within the database. The data management system may perform a pattern matching procedure to evaluate the set of metadata attributes for the structured data within the database against one or more patterns associated with a data type to determine one or more locations within the database that include structured data of the data type. Based on the pattern matching procedure, the data management system may output an indication that the one or more locations within the database include structured data of the data type.