中国民航飞行学院学报

大学生创新竞赛作为高校教育培养体系的重要组成部分,能够促进学生理论和实践相结合,从而提升学生的综合素质。目前,大学生创新竞赛一般采用两阶段评审(网评、现场评审)的方法进行评审。本文首先采用了0-1整数规划模型进行分析设计,确定了最优的“交叉分发”方案。其次,基于BP神经网络讨论比较了T分数法和加权T分数法两种评审方案之间的优劣。最后,针对大学生创新竞赛具有极差大和创新性强的特点,本文提出了相应的改进方案。通过讨论相关指标和实施细节,提出了改进方案,从而保证竞赛成绩的公平性、公正性以及科学性。

As an important component of the cultivation system in the higher education,college student innovation competitions can promote the combination of theory and practice,thereby enhancing students' comprehensive qualities.At present,innovation competitions for college students generally adopt a two-stage review(online review and on-site review).This article firstly uses a 0-1 integer programming model for analysis and design and determines the optimal "cross distribution" scheme.Secondly,this article discusses and compares T-score method and weighted T-score method based on BP neural network,the advantages and disadvantages of two review schemes.Finally,this article proposes corresponding improvement plans in response to the characteristics of large range and strong innovation in college student innovation competitions to ensure the fairness,impartiality and scientificity of competition results.