软集在关联规则挖掘中的应用

发布时间：2018-04-21 01:25

本文选题：软集 + Vague软集　；参考：《西南交通大学》2017年硕士论文

【摘要】：随着信息科学与技术的快速发展,数据库数量不断地增大,而在这些数据中存在各种各样的不确定性问题,如何有效地对这些数字信息进行管理并从中获取所需的知识已经成为当前信息技术研究的热点之一。1999年,Molodtsov提出了软集理论,认为对于复杂事务可以从不同的侧面进行分析,其结果都是对复杂事务的近似刻画,将这些近似刻画综合后可以得到对复杂事务相对精确的描述。软集作为一种新的处理不确定性问题的数学工具,在不确定性决策领域获得了广泛应用。本文主要研究软集在关联规则挖掘中的应用,主要工作如下:一、讨论了 Vague软集排序方法。基于Vague软集的相关运算以及可能度理论,针对Vague软集中对象提出了一种排序方法,进而给出了 Vague软集的一种排序方法,讨论了排序方法的基本性质。通过例子说明了本文提出的对象排序方法以及Vague软集排序方法的有效性。二、讨论了基于软真度的关联规则挖掘方法。基于软集的逻辑公式,提出了一种基于软集的关联规则挖掘方法。将软真度引入软集数据关联规则挖掘,利用软真度描述属性集之间的依赖关系;刻画了软真度与支持度之间的联系,给出了满足给定的支持度阈值和可信度阈值的软关联规则挖掘方法。实例分析结果表明,该方法可约简冗余,提高效率。
[Abstract]:With the rapid development of information science and technology, the number of databases is increasing, and there are various uncertainties in these data. How to manage these digital information effectively and obtain the necessary knowledge from them has become one of the hot topics in information technology research. In 1999, Molodtsov proposed soft set theory, which holds that complex transactions can be analyzed from different aspects. The results are all approximate characterizations of complex transactions, and a relatively accurate description of complex transactions can be obtained by synthesizing these approximations. As a new mathematical tool to deal with uncertain problems, soft sets have been widely used in uncertain decision making field. This paper mainly studies the application of soft set in association rule mining. The main work is as follows: first, the sorting method of Vague soft set is discussed. Based on the correlation operation of Vague soft sets and the possibility degree theory, this paper presents a sort method for Vague soft set objects, then gives a sort method of Vague soft set, and discusses the basic properties of the sorting method. An example is given to illustrate the validity of the proposed method and the Vague soft set sorting method. Secondly, the method of mining association rules based on soft trueness is discussed. Based on the logic formula of soft set, a method of association rule mining based on soft set is proposed. The soft truth degree is introduced into the soft set data association rules mining, and the dependence relation between the attribute set is described by using the soft truth degree, and the relation between the soft truth degree and the support degree is described. A soft association rule mining method satisfying the given support threshold and confidence threshold is presented. The analysis results show that this method can reduce redundancy and improve efficiency.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：O159;TP311.13

【参考文献】