Real-Time Sensitive Data Detection via Keyboard Input Trimming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional data loss prevention systems require time-consuming and resource-intensive file format parsing to detect sensitive data, making it difficult to analyze and prevent data breaches in real-time, especially for new or undisclosed file formats.
Innovation Solution
A format-free sensitive data discrimination method and system that records and trims keyboard input characters in real-time, comparing them to predefined terms to determine the presence of sensitive data without parsing the file format, using a log driving module, pre-filtering module, and sensitive data analyzing module.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional data loss prevention systems use file format parsing to detect sensitive data, then detection accuracy is improved, but analysis time and resource consumption increase significantly
Solution Approach 1:
The system performs preliminary actions by recording keyboard input characters in real-time as they are entered, before the file is fully created or formatted. This allows the sensitive data to be captured at the source during the file generation process, eliminating the need for time-consuming post-creation parsing while maintaining detection accuracy
Solution Approach 2:
The invention extracts only the essential information needed for sensitive data detection - specifically the keyboard input characters - and separates this from the rest of the file processing. By focusing solely on capturing and analyzing the character input data rather than parsing the entire file structure, the system achieves fast detection without sacrificing accuracy
2Adaptability or versatility
If traditional systems parse all file formats to detect sensitive data, then comprehensive detection coverage is achieved, but system complexity and development cost increase
Solution Approach 1:
The system achieves universal applicability across all file formats by capturing sensitive data at the keyboard input level, which is format-agnostic. Since the detection occurs during character entry rather than during file parsing, a single implementation can handle any file type (text, binary, encoded, or undisclosed formats) without requiring separate parsers for each format
Solution Approach 2:
The invention introduces keyboard input recording as an intermediary mechanism between data creation and file storage. This intermediary captures the sensitive information in a standardized format (sequence of characters) that can be analyzed regardless of the eventual file format, eliminating the need for format-specific parsing logic
3Adaptability or versatility
If reverse engineering is used to analyze undisclosed file formats, then detection capability for new formats is improved, but processing load and time consumption remain high
Solution Approach 1:
By recording keyboard inputs during file generation, the system performs the detection action preliminarily, before the file is complete or its format is known. This eliminates the need for resource-intensive reverse engineering later, as the sensitive data has already been captured in its raw character form during the low-cost input phase
Solution Approach 2:
The system creates a simplified copy of the sensitive data information by recording only the keyboard character inputs, rather than attempting to copy or replicate the complex file structure. This lightweight character sequence copy can be analyzed instantly without the heavy processing required to reverse-engineer the original file format
Data Source
AI summary
An exemplary embodiment of the present disclosure illustrates a sensitive data discrimination method executed in a data loss prevention system to determine whether a file has the least one sensitive data during a file generation proceeding. Steps of the sensitive data discrimination method are illustrated as follows. Multiple characters inputted via a keyboard are recorded. The recorded characters are trimmed to generate a trimmed data. The trimmed data and at least one predefined term related to the at least one sensitive data are compared, to determine whether the trimmed data has the at least one sensitive data.