Measure Effectiveness of Change-based Three-way Decision Using Utility Theory

被引:2
作者
Jiang, Chunmao [1 ]
Guo, Doudou [2 ]
Duan, Ying [3 ]
机构
[1] Harbin Normal Univ, Coll Comp Sci & Informat Engineer, Harbin 150025, Heilongjiang, Peoples R China
[2] Southwest Univ, Coll Artificial Intelligence, Chongqing 400715, Peoples R China
[3] Univ Regina, Dept Comp Sci, Regina, SK S4S 0A2, Canada
关键词
Three-way decision; Trisecting-and-acting model; Outcome measure; Utility theory; PROBABILISTIC ROUGH SET; ATTRIBUTE REDUCTION; CONFLICT-ANALYSIS;
D O I
10.1007/s12559-022-09999-x
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The trisecting-acting-outcome (TAO) model of three-way decision plays a crucial role in solving complex problems. It comprises three components: trisecting, acting, and outcome. Trisecting means dividing an entire set into three relatively small parts based on an evaluation function, and acting means designing the corresponding action to apply to these three parts. The outcome denotes evaluation effectiveness based on the first two steps. Several previous investigations have measured effects from a coarse-grained perspective. Still, they have not considered the impact of fine-grained changes in objects and have not devised the method of selecting a strategy-constructed tripartition for a given expected outcome. The present study aimed to explore the outcome measure from the view of fine-grained changes in objects named change-based three-way decision. It also provides a novel approach to quantifying outcomes using utility theory. Specifically, we conducted the study using two perspectives: top-down and bottom-up, depending on the decision demands. The object is first divided into three regions based on changes corresponding to different utilities from the top-down view. Then, the overall utility is aggregated based on the particular utility metrics representing different satisfaction levels in the three regions. From the bottom-up perspective, we derive a pair of thresholds based on the desired utility of the decision-maker and thus obtain three parts whose changes can make by some action to achieve these effects. The principal conclusions of this exploration are that outcomes can be better ranked using objects' changes using utility theory and that strategies can be better acting to achieve corresponding variations based on the expected utility of the decision-maker. We describe the specific steps and procedures of the proposed method through several examples. The findings presented in this paper add to our understanding of the outcome measure for the three-way decision.
引用
收藏
页码:1009 / 1018
页数:10
相关论文
共 40 条
  • [1] Three-way decision support for diagnosis on focal liver lesions
    Chen, Yufei
    Yue, Xiaodong
    Fujita, Hamido
    Fu, Siyuan
    [J]. KNOWLEDGE-BASED SYSTEMS, 2017, 127 : 85 - 99
  • [2] Attribute-oriented cognitive concept learning strategy: a multi-level method
    Fan, Bingjiao
    Tsang, Eric C. C.
    Xu, Weihua
    Chen, Degang
    Li, Wentao
    [J]. INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS, 2019, 10 (09) : 2421 - 2437
  • [3] Ganter B., 2012, FORMAL CONCEPT ANAL
  • [4] A Novel TODIM Method-Based Three-Way Decision Model for Medical Treatment Selection
    Hu, Junhua
    Yang, Yao
    Chen, Xiaohong
    [J]. INTERNATIONAL JOURNAL OF FUZZY SYSTEMS, 2018, 20 (04) : 1240 - 1255
  • [5] Jia X.Y., 2014, LNCS Transactions on Rough Sets, VXVIII, P69, DOI DOI 10.1007/978-3-662-44680-5_5
  • [6] Three-way decisions based feature fusion for Chinese irony detection
    Jia, Xiuyi
    Deng, Zhao
    Min, Fan
    Liu, Dun
    [J]. INTERNATIONAL JOURNAL OF APPROXIMATE REASONING, 2019, 113 : 324 - 335
  • [7] Adaptive thresholds determination for saving cloud energy using three-way decisions
    Jiang, Chunmao
    Wu, Junwei
    Li, Zhicong
    [J]. CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS, 2019, 22 (Suppl 4): : S8475 - S8482
  • [8] Resource-utilization-aware task scheduling in cloud platform using three-way clustering
    Jiang, Chunmao
    Duan, Ying
    Yao, Jun
    [J]. JOURNAL OF INTELLIGENT & FUZZY SYSTEMS, 2019, 37 (04) : 5297 - 5305
  • [9] Effectiveness measures in movement-based three-way decisions
    Jiang, Chunmao
    Yao, Yiyu
    [J]. KNOWLEDGE-BASED SYSTEMS, 2018, 160 : 136 - 143
  • [10] Approximate concept construction with three-way decisions and attribute reduction in incomplete contexts
    Li, Meizheng
    Wang, Guoyin
    [J]. KNOWLEDGE-BASED SYSTEMS, 2016, 91 : 165 - 178