User Behavior Analysis in Privacy Protection with Large Language Models: A Study on Privacy Preferences with Limited Data

2025-05-08

AI Summary

To address the challenges of modeling user privacy preferences and ensuring robust privacy protection under data-scarce conditions, this paper proposes a fine-grained modeling framework that integrates large language models (LLMs) with privacy-enhancing technologies. Methodologically, it pioneers the combination of few-shot learning and privacy-preserving computation to construct an LLM-based privacy behavior understanding module; incorporates differential privacy mechanisms and a federated learning architecture to enable collaborative modeling with minimal data dependency; and employs anonymization and synthetic data augmentation to improve generalizability. Experimental results demonstrate that the proposed approach significantly improves privacy preference prediction accuracy under limited real-world data, while reducing the risk of user data leakage by 62% compared to conventional methods—achieving both high predictive accuracy and strong privacy guarantees.

Technology Category

• Machine Learning: Privacy
• Natural Language Processing: (Large) Language Models
• Humans and AI: Learning Human Values and Preferences

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Security and Privacy: Privacy-enhancing technologies
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

With the widespread application of large language models (LLMs), user privacy protection has become a significant research topic. Existing privacy preference modeling methods often rely on large-scale user data, making effective privacy preference analysis challenging in data-limited environments. This study explores how LLMs can analyze user behavior related to privacy protection in scenarios with limited data and proposes a method that integrates Few-shot Learning and Privacy Computing to model user privacy preferences. The research utilizes anonymized user privacy settings data, survey responses, and simulated data, comparing the performance of traditional modeling approaches with LLM-based methods. Experimental results demonstrate that, even with limited data, LLMs significantly improve the accuracy of privacy preference modeling. Additionally, incorporating Differential Privacy and Federated Learning further reduces the risk of user data exposure. The findings provide new insights into the application of LLMs in privacy protection and offer theoretical support for advancing privacy computing and user behavior analysis.

Problem

• Analyzing user privacy preferences with limited data using LLMs
• Improving privacy preference modeling accuracy in data-limited scenarios
• Reducing user data exposure risk with differential privacy and federated learning

Innovation

• Uses Few-shot Learning for privacy preference modeling
• Integrates Differential Privacy to reduce data exposure
• Applies Federated Learning for secure user data analysis