数据驱动多粒序贯三支分类模型  

Mode of Data-driven Sequential Three-way Decision Classificationbased on Granular Computing

作　　者：郑颖春[1] 卢昕云 ZHENG Ying-chun;LU Xin-yun(School of Sciences,Xi'an University of Science and Technology,Xian 710054,China)

机构地区：[1]西安科技大学理学院,陕西西安710054

出　　处：《模糊系统与数学》2024年第4期178-192,共15页Fuzzy Systems and Mathematics

基　　金：青年科学基金项目(12001420);陕西省技术创新引导专项(基金),2020CGXNG-013。

摘　　要：面对多粒度决策场景,序贯三支决策模型对其中不确定信息,从较粗的粒度向更细粒度进行挖掘,在不同粒度级别下设置不同决策代价。现有模型更多是主观的设置粒层的阅值,因此会忽略粒层中可被分类的等价类,使得模型效率降低并且适用性较差。针对该问题提出一种数据驱动多粒序贯三支分类模型,首先结合等价类的重要度与候选比例函数得到代价函数矩阵,从而获得数据驱动阈值序列。其次证明了阅值的适用性,并且结合等价类的更新状态降低算法的时间复杂度。最后给出实例说明分类算法流程,选取不同的数据集与现有的序贯三支决策方法进行比较,数值实验表明该模型的可以提高准确率的同时保证较好的运算效率,提高了决策的科学性和合理性。Sequential three-way decision rough set can mine uncertain information in multi-granularity decision scenes from coarser granularity to finer granularity,and set different decision costs at different granularity levels.The existing models are more subjective when setting the threshold of particle layer,so the equivalent classes that can be classified in granularity levels are ignored.This will reduce the efficiency and applicability of the model.To solve this problem,a mode of data driven sequential three-way decision classification based on granular computing is proposed.Firstly,the cost function matrix is obtained by combining the significance of the equivalence class and the candidate proportion function,and then the data-driven threshold sequence is obtained.Secondly,the applicability of the threshold is proved,and the time complexity of the algorithm is reduced by combining the update state of the equivalent class.Finally,an example is given to illustrate the classification algorithm flow,and different UCI data sets are selected to compare with the existing sequential three-way decision method.Numerical experiments show that the model has better accuracy,higher operating efficiency and more scientific and reasonable decision-making.

关 键 词：粗糙集 粒计算 序贯三支决策 代价函数矩阵 粒度更新 

分 类 号：O225[理学—运筹学与控制论] TP391.4[理学—数学]