三支概念的恢复集  

Recovery set of three-way concept

作　　者：陈曜琦 徐伟华 蒋宗颖 CHEN Yaoqi;XU Weihua;JIANG Zongying(School of Artificial Intelligence,Southwest University,Chongqing 400715,China)

机构地区：[1]西南大学人工智能学院,重庆400715

出　　处：《山东大学学报（理学版）》2023年第12期52-62,共11页Journal of Shandong University(Natural Science)

基　　金：国家自然科学基金资助项目(62376229);重庆市研究生科研创新项目(CYS21133)。

摘　　要：将三支形式概念分析这一工具引入到数据恢复领域,通过定义三支概念的恢复集和恢复度,研究三支概念间的隐藏信息,提出了一种有效的形式背景恢复算法。同时,针对三支概念恢复集问题,研究三支概念对形式背景二元关系的约束,设计了恢复集的合取范式化简(conjunctive normal form simplification,CNFS)算法,进一步给出了恢复集的动态更新算法,以适应形式背景的不断变化。最后,使用UCI机器学习数据库中的数据集对CNFS算法进行了测试。实验结果表明,CNFS算法在形式背景恢复方面具有较高的准确性和有效性,同时也验证了不同概念对认知的重要程度是不同的。The theory of three-way formal concept analysis is introduced into the field of data recovery in this paper.By defining the recovery set and recovery degree of three-way concept,it explores the hidden information from three-way concepts,and proposes an effective formal context recovery algorithm.Additionally,to solve the three-way concept recovery set problem,the constraints of three-way concept are considered on formal context binary relations,and a conjunctive normal form simplification algorithm is designed for the recovery set(CNFS).Furthermore,a dynamic update algorithm is provided for the recovery set to adapt to the continuous changes of the formal context.Finally,some numerical experiments on public datasets from the UCI perform the effectiveness of our proposed method.Experimental results indicate that the proposed algorithm has high accuracy and effectiveness in formal context recovery,and also verifies that different concepts have different important degrees in cognition.

关 键 词：合取范式 动态更新 三支概念 形式概念分析 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]