P
Ping Xiong
Researcher at Zhongnan University of Economics and Law
Publications - 30
Citations - 489
Ping Xiong is an academic researcher from Zhongnan University of Economics and Law. The author has contributed to research in topics: Differential privacy & Computer science. The author has an hindex of 8, co-authored 24 publications receiving 379 citations.
Papers
More filters
Journal ArticleDOI
Correlated Differential Privacy: Hiding Information in Non-IID Data Set
TL;DR: Experimental results show that the proposed solution outperforms traditional differential privacy in terms of mean square error on large group of queries, which suggests the correlated differential privacy can successfully retain the utility while preserving the privacy.
Journal ArticleDOI
An effective privacy preserving algorithm for neighborhood-based collaborative filtering
TL;DR: The results from experiments show that the proposed PNCF algorithm can obtain a rigid privacy guarantee without high accuracy loss, and the proposed algorithm can resist a KNN attack while retaining the accuracy of recommendations.
Proceedings ArticleDOI
Differential privacy for neighborhood-based collaborative filtering
TL;DR: This study proposes a Private Neighbor Collaborative Filtering algorithm that includes two privacy-preserving operations: Private Neighbor Selection and Recommendation-Aware Sensitivity.
Journal ArticleDOI
Android malware detection with contrasting permission patterns
TL;DR: The contrasting permission patterns are introduced to characterize the essential differences between malwares and clean applications from the permission aspect and a framework based on contrasting permission pattern is presented for Android malware detection.
Journal ArticleDOI
Privacy-preserving topic model for tagging recommender systems
TL;DR: This paper proposes a privacy- preserving tag release algorithm, PriTop, designed to satisfy differential privacy, a strict privacy notion with the goal of protecting users in a tagging dataset, and presents extensive experimental results on four real-world datasets.