Abstract
This study proposes a dynamic adjustment mechanism for evaluating privacy budget parameters to address privacy protection challenges in multi-turn interactions of generative chatbots. By classifying queries across multiple dimensions - including domain, sensitivity, and risk - we construct a comprehensive privacy-preserving framework. Traditional differentially private training methods typically allocate a fixed privacy budget under the assumption of static query behaviors, which often fails to reflect the dynamic nature of real-world interactions. To overcome this limitation, we introduce a query-adaptive mechanism that adjusts the privacy budget per query based on characteristics such as sensitivity, frequency, and temporal patterns. This approach enables more precise control of cumulative privacy loss and mitigates the risks of repeated or excessive querying. Simulation results show that our method achieves an effective balance between privacy preservation and response quality across various query frequencies, domains, and scenarios. This study offers an innovative solution for privacy-preserving generative chatbots and contributes to both the theoretical and practical development of differential privacy in interactive applications.
| Original language | English |
|---|---|
| Title of host publication | International Conference on Electrical, Computer, and Energy Technologies, ICECET 2025 |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| ISBN (Electronic) | 9798331535599 |
| DOIs | |
| State | Published - 2025 |
| Event | IEEE International Conference on Electrical, Computer and Energy Technologies, ICECET 2025 - Paris, France Duration: 03 07 2025 → 06 07 2025 |
Publication series
| Name | International Conference on Electrical, Computer, and Energy Technologies, ICECET 2025 |
|---|
Conference
| Conference | IEEE International Conference on Electrical, Computer and Energy Technologies, ICECET 2025 |
|---|---|
| Country/Territory | France |
| City | Paris |
| Period | 03/07/25 → 06/07/25 |
Bibliographical note
Publisher Copyright:© 2025 IEEE.
Keywords
- Adaptive Privacy Protection
- Differential Privacy
- Dynamic Privacy Budget
- Generative Chatbot
Fingerprint
Dive into the research topics of 'Dynamic Privacy Budget Allocation Mechanism Based on Multi-Dimensional Query Management'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver