An Order-Theoretic Study on Formal Concept Analysis

Author:

Syau Yu-Ru1,Lin En-Bing2,Liau Churn-Jung3ORCID

Affiliation:

1. Department of Information Management, National Formosa University, Yunlin 63201, Taiwan

2. School of Computing and Data Science, Wentworth Institute of Technology, Boston, MA 02115-5998, USA

3. Institute of Information Science, Academia Sinica, Taipei 11529, Taiwan

Abstract

This paper employs an order-theoretic framework to explore the intricacies of formal concepts. Initially, we establish a natural correspondence among formal contexts, preorders, and the resulting partially ordered sets (posets). Leveraging this foundation, we provide insightful characterizations of atoms and coatoms within finite concept lattices, drawing upon object intents. Expanding from the induced poset originating from a formal context, we extend these characterizations to discern join-irreducible and meet-irreducible elements within finite concept lattices. Contrary to a longstanding misunderstanding, our analysis reveals that not all object and attribute concepts are irreducible. This revelation challenges the conventional belief that rough approximations, grounded in irreducible concepts, offer sufficient coverage. Motivated by this realization, the paper introduces a novel concept: rough conceptual approximations. Unlike the conventional definition of object equivalence classes in Pawlakian approximation spaces, we redefine them by tapping into the extent of an object concept. Demonstrating their equivalence, we establish that rough conceptual approximations align seamlessly with approximation operators in the generalized approximation space associated with the preorder corresponding to a formal context. To illustrate the practical implications of these theoretical findings, we present concrete examples. Furthermore, we delve into the significance and potential applications of our proposed rough conceptual approximations, shedding light on their utility in real-world scenarios.

Funder

Taiwan’s NSTC

Publisher

MDPI AG

Subject

Geometry and Topology,Logic,Mathematical Physics,Algebra and Number Theory,Analysis

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3