# Efficient Data Perturbation for Privacy Preserving and Accurate Data Stream Mining

arXiv: https://arxiv.org/abs/1806.06151 · HTML: https://mlsanity.com/paper/1806.06151/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Jun 2018.

## Built on by (most cited first, 12 of 12)

- [A Model or 603 Exemplars: Towards Memory-Efficient Class-Incremental Learning](https://mlsanity.com/paper/2205.13218/) (2022)
- [A Survey on Vulnerability of Federated Learning: A Learning Algorithm Perspective](https://mlsanity.com/paper/2311.16065/) (2023)
- [DiffusePast: Diffusion-based Generative Replay for Class Incremental Semantic Segmentation](https://mlsanity.com/paper/2308.01127/) (2023)
- [Adaptive Adapter Routing for Long-Tailed Class-Incremental Learning](https://mlsanity.com/paper/2409.07446/) (2024)
- [Gradient Leakage Defense with Key-Lock Module for Federated Learning](https://mlsanity.com/paper/2305.04095/) (2023)
- [Securing the Classification of COVID-19 in Chest X-ray Images: A Privacy-Preserving Deep Learning Approach](https://mlsanity.com/paper/2203.07728/) (2022)
- [An Efficient Replay for Class-Incremental Learning with Pre-trained Models](https://mlsanity.com/paper/2408.08084/) (2024)
- [Gradients Stand-in for Defending Deep Leakage in Federated Learning](https://mlsanity.com/paper/2410.08734/) (2024)
- [TACLE: Task and Class-aware Exemplar-free Semi-supervised Class Incremental Learning](https://mlsanity.com/paper/2407.08041/) (2024)
- [HydraGAN A Multi-head, Multi-objective Approach to Synthetic Data Generation](https://mlsanity.com/paper/2111.07015/) (2021)
- [AggSS: An Aggregated Self-Supervised Approach for Class-Incremental Learning](https://mlsanity.com/paper/2408.04347/) (2024)
- [DRL: Discriminative Representation Learning with Parallel Adapters for Class Incremental Learning](https://mlsanity.com/paper/2510.12107/) (2025)

## Abstract (arXiv)

The widespread use of the Internet of Things (IoT) has raised many concerns, including the protection of private information. Existing privacy preservation methods cannot provide a good balance between data utility and privacy, and also have problems with efficiency and scalability. This paper proposes an efficient data stream perturbation method (named as P2RoCAl). P2RoCAl offers better data utility than similar methods: classification accuracies of P2RoCAl perturbed data streams are very close to those of the original data streams. P2RoCAl also provides higher resilience against data reconstruction attacks.