支持向量机在高校教学质量评价中的应用研究

Application of University's Teaching Quality Evaluation Based on Support Vector Machine

研究准确评价教学质量,有助于实现教学科学管理。高校教学质量评价是一个多层次、多目标的系统工程,传统评估方法存在计算繁琐、主观性大等缺陷。为了提高教学质量评价的准确率,在分析多种教学质量评价模型的基础上,提出改进支持向量机的高校教学质量评价模型。将教学质量评价指标量化成后作为支持向量机输入,评价结果作为输出,采用支持向量机非线性学习能力逼近输入与输出之间函数关系,并采用遗传算法对评价模型参数进行优化。仿真结果表明,改进支持向量机提高了教学质量的评价准确率,克服专家系统的主观因素和神经网络的泛化能力差的缺陷,改进方法为高校教学质量的评价提供了参考依据。

Teaching quality evaluation helps realizing teaching scientific management.University teaching quality assessment is a system engineering with multi-level and multi-objective,and the traditional evaluation methods have the defects of trivial calculation and subjectivity.In order to improve the evaluation results of university teaching quality,a university teaching quality evaluation model was proposed based on improved support vector machine by the analysis of many teaching quality evaluation models.Teaching quality evaluation index was quantified as a support vector machine input,and the evaluation result as the output.Using the nonlinear learning ability of support vector machine,the relationship between input and output was approximated,and genetic algorithm was adopted to optimize the evaluation model parameters.Simulation results show that the improved support vector machine can improve the teaching quality evaluation accuracy and overcome the subjective factors of expert system and the generalization ability of neural network.It is suitable for university teaching quality evaluation.