This paper presents a general survey and comparison of algorithms for association rule mining. Association rule mining is a popular pattern discovery method in KDD, involving the identification of rules of the form X ⇒ Y, where X and Y are sets of items. The goal is to find rules that are frequently occurring in a database, with a minimum support and confidence threshold. The main challenges in association rule mining are the exponential growth of potential rules and the need to efficiently prune the search space. The paper discusses the fundamental principles of association rule mining, including the support and confidence measures. It introduces a general framework for association rule mining and describes the basic strategies used in current algorithms. The paper then presents a systematic comparison of the most common algorithms, including Apriori, DIC, Partition, and Eclat. These algorithms differ in their approaches to traversing the search space and determining the support values of itemsets. The paper compares the performance of these algorithms based on runtime experiments and theoretical considerations. It finds that the runtime behavior of the algorithms is more similar than expected, with no single algorithm fundamentally outperforming the others. The paper also discusses various optimizations, such as the use of prefix-trees and fast intersections, which help reduce the computational overhead of association rule mining. The results show that while some algorithms perform better under certain conditions, the overall performance differences are relatively small. The paper concludes that association rule mining algorithms are highly dependent on the data characteristics and the specific requirements of the application. The results suggest that the advantages and disadvantages of different strategies for determining the support values of frequent itemsets nearly balance out on market basket-like data. The paper also highlights the importance of considering both the algorithmic efficiency and the practical applicability of association rule mining techniques.This paper presents a general survey and comparison of algorithms for association rule mining. Association rule mining is a popular pattern discovery method in KDD, involving the identification of rules of the form X ⇒ Y, where X and Y are sets of items. The goal is to find rules that are frequently occurring in a database, with a minimum support and confidence threshold. The main challenges in association rule mining are the exponential growth of potential rules and the need to efficiently prune the search space. The paper discusses the fundamental principles of association rule mining, including the support and confidence measures. It introduces a general framework for association rule mining and describes the basic strategies used in current algorithms. The paper then presents a systematic comparison of the most common algorithms, including Apriori, DIC, Partition, and Eclat. These algorithms differ in their approaches to traversing the search space and determining the support values of itemsets. The paper compares the performance of these algorithms based on runtime experiments and theoretical considerations. It finds that the runtime behavior of the algorithms is more similar than expected, with no single algorithm fundamentally outperforming the others. The paper also discusses various optimizations, such as the use of prefix-trees and fast intersections, which help reduce the computational overhead of association rule mining. The results show that while some algorithms perform better under certain conditions, the overall performance differences are relatively small. The paper concludes that association rule mining algorithms are highly dependent on the data characteristics and the specific requirements of the application. The results suggest that the advantages and disadvantages of different strategies for determining the support values of frequent itemsets nearly balance out on market basket-like data. The paper also highlights the importance of considering both the algorithmic efficiency and the practical applicability of association rule mining techniques.