With the onset of the Information Era and the rapid growth of information technology, ample space for processing and extracting data has opened up. However, privacy concerns may stifle expansion throughout this area. The challenge of reliable mining techniques when transactions disperse across sources is addressed in this study. This work looks at the prospect of creating a new set of three algorithms that can obtain maximum privacy, data utility, and time savings while doing so. This paper proposes a unique double encryption and Transaction Splitter approach to alter the database to optimize the data utility and confidentiality tradeoff in the preparation phase. This paper presents a customized apriori approach for the mining process, which does not examine the entire database to estimate the support for each attribute. Existing distributed data solutions have a high encryption complexity and an insufficient specification of many participants' properties. Proposed solutions provide increased privacy protection against a variety of attack models. Furthermore, in terms of communication cycles and processing complexity, it is much simpler and quicker. Proposed work tests on top of a realworld transaction database demonstrate that the aim of the proposed method is realistic.
翻译:随着信息时代的到来和信息技术的快速发展,已经为处理和提取数据开辟了充足的空间。然而,隐私问题可能抑制了该领域的扩张。本研究针对交易分散在多个源中的可靠挖掘技术的挑战进行探讨。本文考虑创建一组新的算法,可以在实现最大隐私、数据实用性和时间节省等方面进行优化。本文提出了一种独特的双重加密和事务分离器方法,以优化数据库变化来平衡数据实用性和保密性的权衡。本文对于挖掘过程提出了一种定制的Apriori方法,该方法不需要检查整个数据库以估计每个属性的支持度。现有的分布式数据解决方案具有较高的加密复杂度和多个参与者属性的不足规范。所提出的解决方案提供了对多种攻击模型的增强隐私保护。此外,它在通信周期和处理复杂度方面要简单得多且更快速。在真实交易数据库的基础上进行的测试表明,拟议的方法是切实可行的。