共查询到16条相似文献,搜索用时 62 毫秒
1.
针对区间值信息系统,提出一种新的优势关系,并定义了基于这种优势关系的信息系统的上、下近似集。通过可辨识矩阵的方法,提出了区间值信息系统的属性约简方法。通过实例验证该属性约简方法简便易行。 相似文献
2.
3.
介绍集值信息系统和区间值信息系统,并提出了同时具有这两种系统特点的区间集值信息系统.依据属性值的语义关系,将区间集值信息系统分为两类:析取(I型)和合取(II型)系统,并对其分别提出了基于优势关系的粗糙集模型,讨论了相关性质.最后用实例分析验证了所提出系统的有效性. 相似文献
4.
5.
《计算机科学与探索》2017,(4):652-658
因信息系统的复杂性和不确定性,对象的属性值难以用精确的数值来表达,而是采用区间形式表示。针对这一问题,对区间值进一步模糊化,并引进优势关系,建立了不协调区间值模糊序决策信息系统。通过分布约简和最大分布约简来简化知识的表达,找出二者之间的关系,得到了分布约简和最大分布约简的判定定理以及可辨识属性集和可辨识矩阵;提供了不协调的区间值模糊序信息系统的分布约简和最大分布约简的具体方法;结合投资风险这一具体案例的求解分析,进一步阐述了对分布约简研究的意义,丰富了区间值模糊序决策信息系统中的粗糙集方法。 相似文献
6.
7.
8.
9.
考虑到模糊信息系统的不完备性和信息值的不确定性,讨论了不完备区间值模糊信息系统的粗糙集理论,给出了粗糙近似算子的性质。研究了不完备区间值模糊信息系统上的知识发现,提出了基于不完备区间值决策表的决策规则和属性约简,最后给出算例。 相似文献
10.
粗糙集理论与概率论、模糊数学和证据理论等其他处理不确定或不精确问题的理论有很强的互补性。在优势关系的基础上,以证据理论中的mass函数为基本工具,提出了基于优势关系的随机信息系统,研究了优势关系下随机信息系统的属性约简问题。分别考虑了随机信息系统和目标随机信息系统两种情况,并给出了实例说明约简方法的有效性。 相似文献
11.
Lihua Wei Zhenmin Tang Runyun Wang Xibei Yang 《Automatic Control and Computer Sciences》2008,42(5):255-263
As one of the useful extensions of classical rough set approach, the dominance-based rough set approach has been successfully
applied into multi-criteria decision problems. However, the traditional dominance-based rough set approach is only suitable
for the condition attributes, which are positively related with classification analysis. To solve this problem, we propose
an extension of the dominance-based rough set approach in incomplete information system by assuming the condition attributes,
which are not only positively but also negatively related with classification analysis. Furthermore, by considering the existence
of unknown values in incomplete information system, we present the concept of valued dominance relation, which shows the probability
of an object is dominating another one with respect to the condition attributes. By using the valued dominance relation, the
fuzzy dominance-based rough set models are also studied. A numerical example is employed to substantiate the conceptual arguments.
The text was submitted in English by the authors. 相似文献
12.
Since preference order is a crucial feature of data concerning decision situations, the classical rough set model has been generalized by replacing the indiscernibility relation with a dominance relation. The purpose of this paper is to further investigate the dominance-based rough set in incomplete interval-valued information system, which contains both incomplete and imprecise evaluations of objects. By considering three types of unknown values in the incomplete interval-valued information system, a data complement method is used to transform the incomplete interval-valued information system into a traditional one. To generate the optimal decision rules from the incomplete interval-valued decision system, six types of relative reducts are proposed. Not only the relationships between these reducts but also the practical approaches to compute these reducts are then investigated. Some numerical examples are employed to substantiate the conceptual arguments. 相似文献
13.
区间直觉模糊信息系统比一般信息系统更能全面、细致、直观地描述和刻画决策信息,对其进行不确定性研究具有重要的意义。利用信息粒度对区间直觉模糊信息系统的不确定性进行了刻画,给出了区间直觉模糊粒度结构的交、并、差、补等四种运算。提出了区间直觉模糊粒度结构上的三种偏序关系,并建立了它们之间的联系。定义了区间直觉模糊信息粒度和区间直觉模糊信息粒度的公理化,并研究它们的性质。 相似文献
14.
Interval-valued information system is used to transform the conventional dataset into the interval-valued form. To conduct the interval-valued data mining, we conduct two investigations: (1) construct the interval-valued information system, and (2) conduct the interval-valued knowledge discovery. In constructing the interval-valued information system, we first make the paired attributes in the database discovered, and then, make them stored in the neighbour locations in a common database and regard them as ‘one’ new field. In conducting the interval-valued knowledge discovery, we utilise some related priori knowledge and regard the priori knowledge as the control objectives; and design an approximate closed-loop control mining system. On the implemented experimental platform (prototype), we conduct the corresponding experiments and compare the proposed algorithms with several typical algorithms, such as the Apriori algorithm, the FP-growth algorithm and the CLOSE+ algorithm. The experimental results show that the interval-valued information system method is more effective than the conventional algorithms in discovering interval-valued patterns. 相似文献
15.
16.
赵兹 《计算机工程与应用》2017,53(10):38-42
区间值信息系统是常见的信息系统之一。针对相同论域上的区间值信息系统给出属性合成的区间值信息系统;将形式背景的基本理论引入到合成区间值信息系统中,讨论利用子形式背景直接构造原形式背景;进而研究子形式背景与原形式背景下协调集之间的关系,给出原形式背景下同构概念格的构造方法;最后利用实例验证相关结论。 相似文献