Data Loss Prevention for Text Fields Using Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional data loss prevention (DLP) systems face challenges in efficiently scanning network traffic due to high volumes of data, leading to resource strain and increased false positives, as they scan all data including irrelevant information.
Innovation Solution
Implementing a system that identifies and analyzes text fields within forms to determine user-generated content, thereby selectively performing DLP scans only on those fields, reducing unnecessary resource usage and false positives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional DLP systems scan all network traffic through proxies, then data loss prevention coverage is comprehensive, but computing resources are strained and false positives increase
Solution Approach 1:
The patent extracts only the relevant portion (text fields likely containing user-generated content) from the entire network traffic for DLP scanning. By identifying and isolating these specific fields based on form characteristics and content analysis, the system performs scans only on extracted relevant data rather than all traffic, reducing computational load while maintaining detection accuracy.
Solution Approach 2:
The patent applies different scanning quality levels to different parts of the data. Text fields are scanned with full DLP analysis while other fields (images, documents, irrelevant text) are excluded from scanning. This local differentiation allows comprehensive security coverage where needed while optimizing resources away from low-risk areas.
2Reliability
If DLP systems scan all data including irrelevant information, then no data is missed, but computing resources are wasted and false positives increase
Solution Approach 1:
The patent performs preliminary analysis of form characteristics and text field properties before full DLP scanning. By pre-identifying which text fields are likely to contain user-generated content based on form structure, field patterns, and content heuristics, the system prepares and selects only the relevant fields for subsequent scanning, avoiding wasted computational resources on irrelevant data.
Solution Approach 2:
The patent applies partial scanning action by focusing DLP analysis only on the subset of text fields that meet criteria for containing user-generated content. Rather than scanning all data fields exhaustively, the system identifies and scans only the necessary partial set, reducing overall computational consumption while maintaining adequate inspection coverage.
Data Source
AI summary
A computer-implemented method for data loss prevention for text fields may include (1) identifying a form submission sent from a client system, the form submission including a textual field, (2) storing at least one characteristic of a value of the textual field within the form submission in connection with an identifier of a form used to generate the form submission, (3) determining, based at least in part on the characteristic of the value of the textual field, that the textual field may include user-generated content, (4) intercepting a subsequent form submission derived from the form, (5) subjecting the textual field within the subsequent form submission to a data-loss-prevention analysis based at least in part on determining that the textual field may include user-generated content and (6) performing a data-loss-prevention action based on the data-loss-prevention analysis.