Journal of Frontiers of Computer Science and Technology ›› 2022, Vol. 16 ›› Issue (10): 2357-2364.DOI: 10.3778/j.issn.1673-9418.2104118

• Artificial Intelligence • Previous Articles     Next Articles

Research of Inference Rules on Decision Implication and Variable Decision Implication

XUE Dongqian1, ZHAI Yanhui1,2,+(), ZHANG Shaoxia1, LI Deyu1,2, XU Weihua3   

  1. 1. School of Computer and Information Technology, Shanxi University, Taiyuan 030006, China
    2. Key Laboratory of Computational Intelligence and Chinese Information Processing of Ministry of Education, Shanxi University, Taiyuan 030006, China
    3. School of Artificial Intelligence, Southwest University, Chongqing 400715, China
  • Received:2021-04-12 Revised:2021-05-25 Online:2022-10-01 Published:2021-05-31
  • About author:XUE Dongqian, born in 1997, M.S. His research interests include data mining and intelligent decision.
    ZHAI Yanhui, born in 1981, associate professor, Ph.D. supervisor, member of CCF. His research interests include concept lattice and rough set.
    ZHANG Shaoxia, born in 1991, Ph.D. Her research interest is concept lattice.
    LI Deyu, born in 1965, professor, Ph.D. supervisor, senior member of CCF. His research interests include rough set, multi-label learning, etc.
    XU Weihua, born in 1979, professor, Ph.D. supervisor. His research interests include rough set, concept lattice, etc.
  • Supported by:
    National Natural Science Foundation of China(61972238);National Natural Science Foundation of China(62072294);National Natural Science Foundation of China(61806116)

决策蕴涵与可变决策蕴涵上的推理规则研究

薛东前1, 翟岩慧1,2,+(), 张少霞1, 李德玉1,2, 徐伟华3   

  1. 1.山西大学 计算机与信息技术学院,太原 030006
    2.山西大学 计算智能与中文信息处理教育部重点实验室,太原 030006
    3.西南大学 人工智能学院,重庆 400715
  • 通讯作者: + E-mail: chai_yanhui@163.com
  • 作者简介:薛东前(1997—),男,山西运城人,硕士,主要研究方向为数据挖掘、智能决策。
    翟岩慧(1981—),男,山西文水人,副教授,博士生导师,CCF会员,主要研究方向为概念格、粗糙集。
    张少霞(1991—),女,山西浑源人,博士,主要研究方向为概念格。
    李德玉(1965—),男,山西临汾人,教授,博士生导师,CCF高级会员,主要研究方向为粗糙集、多标签学习等。
    徐伟华(1979—),男,山西浑源人,教授,博士生导师,主要研究方向为粗糙集、概念格等。
  • 基金资助:
    国家自然科学基金(61972238);国家自然科学基金(62072294);国家自然科学基金(61806116)

Abstract:

Formal concept analysis is a data analysis method based on concept lattice, while decision implication and variable decision implication are two decision knowledge representations of formal concept analysis. Both decision implication and variable decision implication have complete inference rules for knowledge reasoning. In order to further clarify the differences between variable decision implication and decision implication, this paper first eliminates uncertainty from variable decision implication and introduces restricted variable decision impli-cation. Then, based on the related research on variable decision implication, this paper presents semantic and syntactic characteristics of restricted variable decision implication. On this basis, a complete set of inference rules is obtained for restricted variable decision implication, and decision implication and restricted variable decision implication are compared from the syntactic perspective. Besides, this paper further discusses the differences between the two decision types in knowledge representation and reasoning from the semantic perspective, and summarizes the characteristics of restricted variable decision implication relative to decision implication from the semantic and syntactic perspectives. These conclusions provide new ideas for comparative research of different decision types such as decision implication, variable decision implication, and fuzzy decision implication, as well as a new perspective for introducing more efficient and concise inference rules.

Key words: formal concept analysis, variable decision implication, inference rules, knowledge reasoning ability, knowledge representation

摘要:

形式概念分析是一种基于概念格的数据分析方法,决策蕴涵和可变决策蕴涵是形式概念分析的两种决策知识表示类型。决策蕴涵和可变决策蕴涵上已有完备的推理规则集。为了进一步厘清可变决策蕴涵和决策蕴涵在推理规则上的区别,首先对可变决策蕴涵进行不确定性剥离并提出了受限可变决策蕴涵,然后基于可变决策蕴涵的相关研究给出了受限可变决策蕴涵的语义特征和语构特征;在此基础上,得出了受限决策蕴涵的完备推理规则集,并且从语构角度比较了受限可变决策蕴涵和决策蕴涵;基于此,从语义角度得出了这两种决策类型在知识推理能力和知识表示形式上的差异,并从语义和语构角度总结了受限可变决策蕴涵相对于决策蕴涵所具有的特征。该研究为决策蕴涵、可变决策蕴涵和模糊决策蕴涵等不同决策类型的比较性研究提供了新的思路,为决策蕴涵更为高效、简洁的推理规则提供新的视角。

关键词: 形式概念分析, 可变决策蕴涵, 推理规则, 知识推理能力, 知识表示形式

CLC Number: