单边区间集模糊半概念的知识表示及提取方法

PDF下载HTML

Title:

Knowledge representation and extraction method of partially-known fuzzy semiconcepts

作者:

毛华^1,2, 胥德华^1,2, 刘川¹, 郑博雅¹, 王刚³, 张植明¹

1. 河北大学 数学与信息科学学院, 河北 保定 071002;
2. 河北大学 河北省机器学习与计算机智能重点实验室, 河北 保定 071002;
3. 河北大学 生命科学学院, 河北 保定 071002

Author(s):

MAO Hua^1,2, XU Dehua^1,2, LIU Chuan¹, ZHENG Boya¹, WANG Gang³, ZHANG Zhiming¹

1. College of Mathematics and Information Science, Hebei University, Baoding 071002, China;
2. Hebei Key Laboratory of Machine Learning and Computational Intelligence, Hebei University, Baoding 071002, China;
3. College of Life Science, Hebei University, Baoding 071002, China

关键词:

模糊形式背景; 半概念; 区间集; 区间集概念; 单边区间集概念; 模糊概念; 单边区间集模糊半概念; 知识表示和提取

Keywords:

fuzzy formal context; semiconcept; interval set; interval-set concept; partially-known formal concept; fuzzy concept; partially-known fuzzy semiconcept; knowledge representation and extraction

分类号:

TP18

DOI:

10.11992/tis.202304057

摘要:

为了使“区间”形式加以表述的不确定信息的提取具有侧重性，需提取出对象(属性)集对应的属性(对象)区间集。本文在模糊形式背景中，通过引入2个阈值，将单边区间集与经典半概念结合，提取出属性(对象)集对应的对象(属性)区间集，从而提出区间集外延–集合内涵(集合外延–区间集内涵)(interval set extent-set intent(set extent-interval set intent)，ISE-SI(SE-ISI))型单边区间集模糊半概念。全体ISE-SI(SE-ISI)型单边区间集模糊半概念构成格，并给出基于格搜寻全体ISE-SI(SE-ISI)型单边区间集模糊半概念的算法。通过与已有成果对比，显示出这2种知识表示形式的多方优势。本文所得结果在知识表示及提取方法上具有适用范围广、实际应用强等优点。

Abstract:

To ensure that the extraction of uncertain information expressed in the form of "intervals" is focused, it is necessary to extract the corresponding attribute (object) interval set for the object (attribute) set. By introducing two thresholds from the fuzzy context, the unilateral interval set is combined with the classical semiconcept to extract the corresponding object (attribute) interval set for the attribute(object) set, and thereby the interval set extent-set intent(set extent-interval set intent)(ISE-SI (SE-ISI)) type partially-known fuzzy semiconcept is proposed. All ISE-SI (SE-ISI) type partially-known fuzzy semiconcepts form a lattice, and the algorithm based on lattice search for all ISE-SI (SE-ISI) type partially-known fuzzy semiconcepts is given. The multiple advantages of these two knowledge representation forms are demonstrated by comparing them with existing achievements. The results obtained in this paper have the advantages of broad applicability and strong practical application in knowledge representation and extraction methods.