Yanagi Tomoya, Ikeda Shunnosuke, Sukegawa Noriyoshi, Takano Yuichi
Graduate School of Science and Technology, University of Tsukuba, Tsukuba-shi, Ibaraki, Japan.
Department of Advanced Sciences, Faculty of Science and Engineering, Hosei University, Koganei-shi, Tokyo, Japan.
PLoS One. 2025 Apr 21;20(4):e0319954. doi: 10.1371/journal.pone.0319954. eCollection 2025.
In order to provide high-quality recommendations for users, it is desirable to share and integrate multiple datasets held by different parties. However, when sharing such distributed datasets, we need to protect personal and confidential information contained in the datasets. To this end, we establish a framework for privacy-preserving recommender systems using the data collaboration analysis of distributed datasets. Numerical experiments with two public rating datasets demonstrate that our privacy-preserving method for rating prediction can improve the prediction accuracy for distributed datasets. More precisely, compared to the individual analysis in which each party analyzes only its own dataset, our method reduced prediction errors by an average of 4.5% and up to 7.0%. This study opens up new possibilities for privacy-preserving techniques in recommender systems.
为了向用户提供高质量的推荐，共享和整合不同方持有的多个数据集是很有必要的。然而，在共享此类分布式数据集时，我们需要保护数据集中包含的个人和机密信息。为此，我们利用分布式数据集的数据协作分析建立了一个隐私保护推荐系统框架。对两个公共评分数据集进行的数值实验表明，我们用于评分预测的隐私保护方法可以提高分布式数据集的预测准确性。更确切地说，与各方仅分析自己数据集的单独分析相比，我们的方法平均将预测误差降低了4.5%，最多降低了7.0%。这项研究为推荐系统中的隐私保护技术开辟了新的可能性。
