Generalized discriminant classification model in view of Choquet integral

2nd International Symposium on Application of Materials Science and Energy Materials

材料科学;能源学

Dai, Xuezhen^1 ; Hu, Zhenni^1 ; Chang, Zaibin^1

Xi'an Traffic Engineering Institute, Shannxi Xi'an, China^1

关键词: Algorithm for solving;    Analysis of data;    Classification models;    Dimensional spaces;    Discriminant classifications;    Fisher criterion;    Fisher discriminant analysis;    N-dimensional space;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/490/4/042042/pdf
DOI  :  10.1088/1757-899X/490/4/042042

学科分类：材料科学（综合）

来源: IOP

【 摘 要 】

Fisher discriminant analysis is a common technique for data processing. The paper proposes a new generalized discriminant classification model in view of Choquet integral. Firstly, through the deep analysis of data classification with regard to Choquet integral, the weighted axis in n-dimensional space is introduced, and each data point in n-dimensional space is corresponding to a point in (2n-1)-dimensional space according to the given weighted axis. Secondly, using the Fisher criterion, a new classification model in view of Choquet integral is presented. Moreover, the algorithms for solving the classification model are built. Finally, some experimental outcomes illustrate the feasibility of the algorithm for solving classification problems.