崔意强,萨日娜

• 1:内蒙古工业大学机械工程学院

摘要(Abstract)：

针对产品数据具有更新快，存在不一致的特点，提出产品改进差别信息树及属性增量式约简方法。将产品属性集看作信息系统，分成两部分，一部分为一致信息系统，论域和值域保持不变；另一部分为不一致信息系统，将同一组的不一致对象只保留一个。针对不一致信息系统，借助核属性去枝策略和加权属性重要度策略，给出改进差别矩阵，构建改进差别信息树。分析新增对象与原信息系统关系分类，给出差别信息树更新机制，给出增量式属性约简算法，以UCI数据集测试了算法。最后以骨折外固定夹板属性增量式约简为例，验证了所提方法的有效性。

关键词(KeyWords)： 增量式属性约简;核属性去枝策略;加权属性重要度;信息系统简化;差别信息树

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金项目（51765052）;; 内蒙古自然科学基金（2020LH05009）;; 内蒙古自治区直属高校基本科研业务费项目（JY20220282）

作者(Author): 崔意强,萨日娜

DOI: 10.19356/j.cnki.1001-3997.20260403.006

参考文献(References)：

• [1]Chen D,Dong L,Mi J. Incremental mechanism of attribute reduction based on discernible relations for dynamically increasing attribute[J]. Soft Computing,2020,24(1):321-332.
• [2]Ma F,Ding M,Zhang T,et al.Compressed binary discernibility matrix based incremental attribute reduction algorithm for group dynamic data[J].Neurocomputing,2019(344):20-27.
• [3]Feng W,Ma Y.An incremental attribute reduction algorithm based on improved binary distinguishable matrix[J]. Journal of Physics:Conference Series,2020,1544(1):012018.
• [4]Ding W P,Qin T ZH,Shen X J,et al.Parallel incremental efficient attribute reduction algorithm based on attribute tree[J]. Information Sciences,2022(610).
• [5]葛浩，李龙澍，杨传健.基于简化差别矩阵的增量式属性约简[J].四川大学学报（工程科学版），2013,45(1):116-124.(Ge Hao,Li Longshu,Yang Chuanjian.Incremental attribute reduction based on simplified discernibility matrix[J].journal of Sichuan University(Engineering Science Edition),2013,45(1):116-124.)
• [6]杨明，吕静.一种基于C-Tree的属性约简增量式更新算法[J].控制与决策，2012,27(12):1769-1775.(Yang Ming,Lyu Jing.An incremental updating algorithm of attribute reduction based on C-Tree[J].Control and Decision,2012,27(12):1769-1775.)
• [7]Pawlak Z.Rough sets[J].International Journal of Information and Computer Science,1982,11(5):341-356.
• [8]王立业，萨日娜，张利春.考虑属性组合关联的机械产品属性约简方法[J].机械设计与制造，2022(2):6-10+16.(Wang Liye,Sa Ri’na,Zhang Lichun. Attribute reduction method of mechanical products considering attribute combination association[J].Mechanical Design&Manufacturing,2022(2):6-10+16.)
• [9]胡秦斌.一种基于决策信息系统的知识约简算法[J].计算机工程与应用，2013,49(16):133-136.(Hu Qinbin.A knowledge reduction algorithm based on decision information system[J].Computer Engineering and Applications,2013,49(16):133-136.)
• [10]Skowron A,Rauszer C.The Discernibility Matrices and Functions In Information Systems[M].Intelligent Decision Support.Springer,Dordrecht,1992:331-362.
• [11]樊忠，李娟莉.提升机Q制动器故障诊断规则知识融合获取方法[J].机械设计与制造，2019(4):123-126+130.(Fan Zhong,Li Juanli. Knowledge fusion acquisition method of hoist brake fault diagnosis rules[J]. Mechanical Design&Manufacturing,2019(4):123-126+130.)
• [12]杜文辽，陈晓雨，李坤，等.旋转机械故障诊断增量式属性约减新方法[J].机械设计与制造，2015(4):163-165+170.(Du Wenliao,Chen Xiaoyu,Li Kun,et al.A new incremental attribute reduction method for rotating machinery fault diagnosis[J].Mechanical Design&Manufacturing,2015(4):163-165+170.)
• [13]Ming Y,Ping Y.A novel condensing tree structure for rough set feature selection[J].Neurocomputing,2007,71(4).
• [14]蒋瑜.基于改进差别信息树的粗糙集属性约简算法[J].控制与决策，2019,34(6):1253-1258.(Jiang Yu. Rough set attribute reduction algorithm based on improved differential information tree[J].Control and Decision,2019,34(6):1253-1258.)
• [15]胡可云.基于概念格和粗糙集的数据挖掘方法研究[D].北京：清华大学，2001.(Hu Keyun.Research on data mining method based on concept lattice and rough set[D].Beijing:Tsinghua University,2001.)