基于改进Apriori算法的频繁项集挖掘

Mining Frequent Itemsets Based on Improved Apriori Algorithm

摘要: 传统的关联规则挖掘算法有三种，分别是Apriori算法、FP-growth算法和Eclat算法。其中传统的Apriori算法简单易实现，但处理海量数据时耗时巨大且磁盘I/O过高，效率低下。而FP-growth算法虽然快速且高效，但对于内存资源极其不友好，且挖掘过程中出现问题难以追踪。针对Apriori算法与FP-growth算法的优缺点，本文提出了一种基于深度递归与散列技术改进的Apriori算法。该算法基于散列技术与递归思想，将传统算法的遍历次数大幅度降低，且很大程度上减少了磁盘I/O，保证了较低的时延和更多的存储空间，在算法时间和空间复杂度方面进行了一定程度上的优化。既提高了传统Apriori算法的效率，同时也保证了算法的可扩展性。

Abstract: There are three traditional association rule mining algorithms, namely Apriori algorithm, FP-growth algorithm and Eclat algorithm. The traditional Apriori algorithm is simple and easy to implement, but it takes a lot of time to process massive data, high disk I/O and low efficiency. Alt-hough FP-growth algorithm is fast and efficient, it is extremely unfriendly to memory resources, and problems in the mining process are difficult to track. Aiming at the advantages and disadvantages of Apriori algorithm and FP-growth algorithm, this paper proposes an improved Apriori algorithm based on deep recursion and hashing technology. Based on hash technology and recursive ideas, the algorithm greatly reduces the traversal times of the traditional algorithm, greatly reduces disk I/O, ensures lower delay and more storage space, and optimizes the time and space complexity of the algorithm to a certain extent. It not only improves the efficiency of the traditional Apriori algorithm, but also ensures the scalability of the algorithm.