Incremental privacy-preserving association rule mining using negative border

Privacy preserving association rule mining can extract important rules from distributed data with limited privacy breaches. Protecting privacy in incremental maintenance for distributed association rule mining is necessary since data are frequently updated. In privacy preserving data mining, scanning all the distributed data is very costly. This paper proposes a new incremental protocol for privacy preserving association rule mining using negative border concept. The protocol scans old databases at most once, and therefore reducing the I/O time. We also conduct experiments to show the efficiency of our protocol over existing ones.