摘要
随着数据库规模的日益增大,关联规则挖掘需要在挖掘效率、可用性、隐私性及精确性等方面得到提升,需要对传统的关联规则挖掘算法进行更新和改进。在传统的Apriori算法基础上,提出了一种新的在关系数据库中挖掘关联规则的算法。该算法只需扫描一次数据库即可得到频繁项集,并通过非频繁项集来减少候选项集的生成,从而提高了算法的运算效率;此外,该算法将包含敏感数据事务做相关的处理,以达到隐藏包含敏感数据的关联规则。理论分析和实验结果表明,新算法不仅提高了关联规则挖掘的效率,而且还达到了隐藏包含敏感规则的目的。
With the growing size of the database,association rules mining needs to be improved in efficiency,availability,privacy and accuracy.Hence it is needed for the traditional association rule mining algorithm to be updated and improved.In this paper,there is a new association rule mining algorithm based on traditional Apriori algorithm in the relational database.The algorithm only needs to scan the database once you can get frequent itemsets,and through non-frequent itemsets to reduce the generation of candidate items to improve the computing efficiency of the algorithm;At the same time,the algorithm will contain sensitive data to do related services processing in order to achieve hiding association rules that contain sensitive data.Theoretical analysis and experimental results show that the new algorithm not only improves the efficiency of mining association rules,but also hides the association rules that contains sensitive rule.
出处
《辽宁科技大学学报》
CAS
2010年第5期481-484,共4页
Journal of University of Science and Technology Liaoning
基金
辽宁省自然科学基金资助项目(20082189)
关键词
关联规则
隐私保护
挖掘算法
association rules
privacy protecting
mining algorithm
