AI Feedback Anonymization With Similarity and Anonymity Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data privacy regulations limit the retention time of verbatim feedback data, posing challenges for processing large amounts of written text while maintaining data privacy, and existing anonymization techniques are time-consuming and alter the meaning or sentiment of the feedback.
Innovation Solution
A verbatim feedback processing system using a generative AI model to anonymize feedback while preserving meaning, tone, and sentiment, combined with similarity and anonymity validation components to ensure the anonymized feedback meets predefined similarity and anonymity thresholds for long-term retention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional anonymization techniques are used to remove personally identifiable information, then data privacy compliance is improved, but the meaning, tone, or sentiment of the feedback is altered
Solution Approach 1:
The patent replaces traditional mechanical anonymization techniques (regex patterns, keyword replacement, manual redaction) with an AI-based natural language processing system. This substitution enables the system to understand contextual meaning while removing PII, thereby maintaining both privacy compliance and feedback integrity. The AI model analyzes the semantic structure and intent of feedback rather than simply applying pattern-matching rules.
Solution Approach 2:
The patent changes the operational parameters of anonymization from rigid pattern-matching with fixed thresholds to dynamic AI-based decision-making. The system adjusts its processing approach based on the complexity and context of each feedback item, enabling nuanced removal of PII while preserving meaningful content. This parameter change allows the system to handle edge cases and contextual variations that traditional methods cannot accommodate.
2Reliability
If data retention time limits are enforced to comply with privacy regulations, then data privacy compliance is improved, but the ability to process and analyze large amounts of feedback data is reduced
Solution Approach 1:
The patent extracts and removes only the personally identifiable information components from feedback data while retaining the core feedback content. This selective extraction approach transforms the data into an anonymized form that can be stored indefinitely without violating privacy regulations, thereby enabling long-term retention and comprehensive analysis of feedback trends.
Solution Approach 2:
The patent creates anonymized copies of the original feedback data that can be stored and processed without legal restrictions. These copies preserve all analytical value while eliminating privacy risks, allowing the organization to maintain both compliant storage and full analytical capabilities simultaneously.
3Reliability
If anonymization processing is applied to large volumes of feedback data, then data privacy compliance is improved, but processing time and computational resources increase
Solution Approach 1:
The patent implements preliminary action by pre-training AI models on feedback data characteristics and pre-configuring processing pipelines. This preparation enables the system to rapidly anonymize new feedback data without extensive processing overhead. The system also performs preliminary assessments to identify feedback items that require minimal or no anonymization intervention.
Solution Approach 2:
The patent employs periodic action through batch processing and asynchronous anonymization workflows. Instead of processing all feedback data synchronously, the system processes data in manageable batches and continues accepting new feedback while previous batches are being anonymized. This periodic approach maintains compliance while minimizing processing bottlenecks.
Data Source
AI summary
A verbatim feedback processing system utilizes generative Artificial Intelligence (AI) which has been trained to rewrite text in a manner that retains the meaning, tone, sentiment, and the like, while also anonymizing the text by removing personal identifiable information. The anonymized feedback is then processed by a similarity validation component and an anonymity validation component. The similarity validation component determines whether the anonymized feedback is within a predetermined similarity threshold of the original feedback. The anonymity validation component determines whether the anonymized feedback satisfies anonymity requirements. If anonymized feedback satisfies similarity and anonymity requirements, the anonymized feedback is stored in an anonymized feedback database.