计算机集成制造系统 ›› 2020, Vol. 26 ›› Issue (12): 3408-3426.DOI: 10.13196/j.cims.2020.12.023

• 当期目次 • 上一篇    下一篇

融合组合赋权与嵌套集成分类器的服务商评价

刘鹏程1,孙林夫1+,张常有2   

  1. 1.西南交通大学信息科学与技术学院
    2.中国科学院软件研究所
  • 出版日期:2020-12-31 发布日期:2020-12-31
  • 基金资助:
    国家重点研发计划资助项目(2017YFB1400902)。

Evaluation of service station based on combined weight and nested ensemble classifier

  • Online:2020-12-31 Published:2020-12-31
  • Supported by:
    Project supported by the National Key Research and Development Program,China(No.2017YFB1400902).

摘要: 为提升制造企业对服务价值链中服务业务活动的管控能力,针对面向制造企业的服务商业务协同能力评价问题,研究了服务商业务协同能力评价指标体系,提出了基于嵌套集成分类器的服务商业务协同能力评价算法。基于主观组合权重和客观组合权重对服务商业务协同能力进行评价,融合两者评价结果形成初始分级评价结果,利用初始分级评价结果和岭回归集成分类决策算法构建嵌套集成分类器,实现对服务商业务协同能力的精确分级评价。实验表明,该嵌套集成分类器较其他同类集成分类算法具有较高的算法性能。基于真实服务业务数据集的实验结果表明,该嵌套集成分类器在服务商业务协同能力评价应用方面具有更为突出的算法性能。

关键词: 服务价值链, 服务商评价, 岭回归, 特征选择, 嵌套集成分类器, 汽车制造企业

Abstract: To improve management and control capacities of manufactory for service value chain,a collaboration capability evaluation index system was designed for service stations who had constructed collaboration relationship with manufactory.Based on the evaluation index system,a nested ensemble classifier was proposed for evaluation of collaboration capability of service station with manufactory.Based on service business dataset,initial evaluation solutions were calculated by subjective combination weights and objective combination weights.Using the initial evaluation solutions and ridge regression method,the nested ensemble classifier was constructed and used to calculate the final evaluation solutions.Experiments showed that the nested ensemble classifier had a promising result compared with other ensemble algorithms.Especially based on real service business data,experimental results illustrated that the performance of nested ensemble classifier was more prominent than others.

Key words: service value chain, service station evaluation, ridge regression, features election, nested ensemble classifier, automobile manufacturing enterprise

中图分类号: