Abstract

Association Rule Mining (ARM) is one of the fundamental techniques in data mining widely used to discover hidden patterns correlations and associations within large datasets. With the exponential growth of data in various domains such as healthcare finance e-commerce social networks and government databases ARM has become increasingly important for decision-making predictive analytics and knowledge discovery. However the widespread use of ARM raises significant concerns about privacy as the extraction of association rules may inadvertently expose sensitive information about individuals or organizations. This research paper focuses on the study of association rule mining methods that incorporate privacy-preserving mechanisms ensuring that useful patterns can be extracted without compromising confidential data. The study examines classical ARM techniques including Apriori FP-Growth and Eclat and evaluates their extensions with privacy-preserving frameworks such as k-anonymity differentia