On efficiency and data privacy level of association rules mining algorithms within parallel spatial data warehouse

Data privacy becomes one of the most important aspects of Data Mining models designing. So far, most of studies on Data Mining algorithms where focused on efficiency and usability. One of a factor of achieved improvements is using a data warehouse as a source of analyzed data. In our work, we focus on the association rule mining within the parallel spatial data warehouse with data privacy preserving. According to the data model of the data warehouse, we use VPSI algorithm which operates on vertically partitioned data. The data protection mechanism uses a secure intersection of sets. The algorithm is based on the parallel version of Apriori algorithm, but instead of mining data on a transaction level, it operates on aggregated data. In our paper we analyze both aspects, efficiency and data privacy level of the algorithm.