Attribute Selection Method based on Objective Data and Subjective Preferences in MCDM

IAENG Internaitonal journal of computer science

Yang Yu¹  Yi Qu²  Yi Feng²  Xiaofei Ma² 

[1] Agricultural Bank of China Data Center88 Aoni Road, Pudong New AreaShanghai City, China, 200131;Technology PlanningDalian Commodity ExchangeDalian City, China, 116023

关键词: attribute selection;    multi-criteria decision-making (MCDM);    multiobjective optimization;    attribute correlation;   

DOI  :  10.15837/ijccc.2018.3.3188

学科分类：计算机科学（综合）

来源: International Association of Engineers

【 摘 要 】

Decision attributes are important parameters when choosing an alternative in a multiple criteria decision-making (MCDM) problem. In order to select the optimal set of decision attributes, an analysis framework is proposed to illustrate the attribute selection problem. Then a two-step attribute selection procedure is presented based on the framework: In the first step, attributes are filtered by using correlation algorithm. In the second step, a multi-objective optimization model is constructed to screen attributes from the results of the first step. Finally, a case study is given to illustrate and verify this method. The advantage of this method is that both external attribute data and subjective decision preferences are utilized in a sequential procedure. It enhances the reliability of decision attributes and matches the actual decision-making scenarios better.

【 授权许可】

CC BY