-means clustering and kNN classification based on negative databases

Open Webpage

Dongdong Zhao, Xiaoyi Hu, Shengwu Xiong, Jing Tian, Jianwen Xiang, Jing Zhou, Huanhuan Li

Published: 2021, Last Modified: 30 Jul 2025Appl. Soft Comput. 2021EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•An algorithm (named QK-hidden) is designed for generating NDBs in a fine-grained manner.•A method for estimating Euclidean distance upon the NDBs generated by QK-hidden algorithm is proposed.•A privacy-preserving k-means clustering algorithm based on the QK-hidden algorithm is presented.•A privacy-preserving kNN classification algorithm based on the QK-hidden algorithm is presented.