An Effective Association Rule Based Algorithm for Privacy Preserving Frequent Itemset Mining

Abstract

Finding hidden patterns in huge data sets is a technique known as data mining or knowledge discovery. There are several data mining algorithms, each with a unique goal. It is a challenging task to extract significant and unidentified patterns from massive datasets. One of the most well-known data mining algorithms for determining the significant link between the item-sets is association rule mining, which has the capacity to uncover unanticipated data dependencies. The fundamental concept is to determine if the existence of some items strongly indicates the existence of other things in a given database of item sets (such as shopping baskets). The apriori algorithm is a popular method for extracting association rules from datasets. It involves identifying frequently occurring groups of items in transaction data and using them to generate association rules. Association rules are a descriptive data mining technique that can provide useful insights into patterns and trends in the data.In general, the traditional apriori algorithm works well for small datasets, but it may not be efficient for handling large datasets. However, by making a few modifications to the implementation of the apriori algorithm, it is possible to improve its performance for large datasets. In this work, we made some adjustments to the apriori algorithm implementation and were able to achieve better results when working with large datasets.