当前位置: X-MOL 学术Int. J. Approx. Reason. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Rule extraction based on linguistic-valued intuitionistic fuzzy layered concept lattice
International Journal of Approximate Reasoning ( IF 3.2 ) Pub Date : 2021-03-01 , DOI: 10.1016/j.ijar.2020.12.018
Li Zou , Hongmei Lin , Xiaoying Song , Kaihua Feng , Xin Liu

As one of the research tools for data processing and knowledge discovery, concept lattice can effectively extract information. In daily life, due to the ambiguity and uncertainty of the decision environment, different experts may provide different evaluation information according to individual needs which may be expressed with linguistic values. In order to handle these problems, we study the rule extraction method of linguistic-valued intuitionistic fuzzy layered concept lattice. First, we propose a linguistic-valued intuitionistic fuzzy formal decision context based on the intuitionistic fuzzy lattice implication algebra, which can simultaneously process the obtained comparable and incomparable linguistic information from both positive and negative aspects. Furthermore, by setting different linguistic-valued trust degrees, we put forward the linguistic-valued hierarchical concept construction operator. On this basis, a linguistic-valued intuitionistic fuzzy layered concept lattice can be constructed to meet the requirements of different experts at different levels. And then through the relationship between the conditional concept and decision concept in the obtained concept set, we get a rule set of the formal decision context. Finally, we present a rule extraction method to help people make more reasonable decisions, combining the confidence and support degree of linguistic-valued intuitionistic fuzzy decision rules. And a practical example involving individual financial investment decision-making is used to verify the efficiency and applicability of the proposed approach.



中文翻译:

基于语言直觉模糊分层概念格的规则提取

作为数据处理和知识发现的研究工具之一,概念格可以有效地提取信息。在日常生活中,由于决策环境的不确定性和不确定性,不同的专家可能会根据个人需求提供不同的评估信息,这些评估信息可能会以语言价值来表达。为了解决这些问题,我们研究了语言值直觉模糊分层概念格的规则提取方法。首先,我们在直觉模糊格蕴涵代数的基础上提出了一种具有语言价值的直觉模糊形式决策上下文,可以同时从正反两个方面处理所获得的可比,不可比的语言信息。此外,通过设置不同的语言价值信任度,我们提出了具有语言价值的层次概念构造算子。在此基础上,可以构建语言价值的直觉模糊分层概念格,以满足不同层次不同专家的需求。然后通过获得的概念集中条件概念与决策概念之间的关系,得到形式决策上下文的规则集。最后,结合语言价值直觉模糊决策规则的置信度和支持度,提出了一种规则提取方法,可以帮助人们做出更合理的决策。并通过一个涉及个人金融投资决策的实例来验证所提方法的有效性和适用性。可以构造一个语言价值的直觉模糊分层概念格,以满足不同层次上不同专家的需求。然后通过获得的概念集中条件概念与决策概念之间的关系,得到形式决策上下文的规则集。最后,结合语言价值直觉模糊决策规则的置信度和支持度,提出了一种规则提取方法,可以帮助人们做出更合理的决策。并通过一个涉及个人金融投资决策的实例来验证所提方法的有效性和适用性。可以构造一个语言价值的直觉模糊分层概念格,以满足不同层次上不同专家的需求。然后通过获得的概念集中条件概念与决策概念之间的关系,得到形式决策上下文的规则集。最后,结合语言价值直觉模糊决策规则的置信度和支持度,提出了一种规则提取方法,可以帮助人们做出更合理的决策。并通过一个涉及个人金融投资决策的实例来验证所提方法的有效性和适用性。然后通过获得的概念集中条件概念与决策概念之间的关系,得到形式决策上下文的规则集。最后,结合语言价值直觉模糊决策规则的置信度和支持度,提出了一种规则提取方法,可以帮助人们做出更合理的决策。并通过一个涉及个人金融投资决策的实例来验证所提方法的有效性和适用性。然后通过获得的概念集中条件概念与决策概念之间的关系,得到形式决策上下文的规则集。最后,结合语言价值直觉模糊决策规则的置信度和支持度,提出了一种规则提取方法,可以帮助人们做出更合理的决策。并通过一个涉及个人金融投资决策的实例来验证所提方法的有效性和适用性。

更新日期:2021-03-18
down
wechat
bug