摘要
基于某大规模创新类竞赛的相关模拟评审数据,首先,通过讨论两评审阶段成绩与极差的变化,分析现有评审方案各自的优劣;其次,构建极差模型,运用Adaboost算法检验模型的预测效果;最后,针对现有评审方案的不足,对未来评审方案改革提出展望.研究结果表明:两阶段评审和不分阶段评审各有优劣,极差模型预测效果较好.
Based on the relevant simulation review data of a large-scale innovation competi-tion,firstly,the advantages and disadvantages of the existing review schemes are analyzed by discuss-ing the changes of the results and ranges in the two review stages.Secondly,the range model is con-structed and the Adaboost algorithm is used to test the prediction effect of the model.Finally,in view of the shortcomings of the existing evaluation schemes,the future evaluation scheme reform is pro posed.The results show that both two-stage review and non-stage review have their own advantages and disadvantages,and the range model has better prediction effect.
作者
陈齐萌
CHEN Qimeng(School of Economics and Management,Anhui University of Science and Technology,Huainan Anhui 232000,China)
出处
《太原师范学院学报（自然科学版）》
2024年第2期19-27,共9页
Journal of Taiyuan Normal University:Natural Science Edition
作者简介
陈齐萌(2000-),男,安徽准南人,在读硕士研究生,主要从事创新管理研究,E-mail:2050294854@qq.com.
