【摘要】 教师教学水平的评估是教育评估的核心问题之一。本研究将支持向量机多分类方法引入教学水平评估任务之中,利用支持向量机将线性模糊不可分的样本映射到高维空间使之具有线性可分特性,从本质上避开了从归纳到演绎的传统过程,简化了非线性问题的分类过程;结构风险最小化理论保证分割的全局最优化,降低期望风险。该方法充分利用支持向量机的小样本学习的高效性,实现了优秀的学习效果,减少了传统评价方法中的分歧误差和主观性因素的影响,更加符合宏观取向的评价结论。该研究成果可与信息熵、模糊数学等研究方法相结合,进一步增强数据拟合的精度,该方法对改进教师的教学水平、促进教学质量的提高具有一定的参考意义。更多还原

【Abstract】 Teaching level evaluation is an important part of educational evaluation.In this study,Support Vector Machine(SVM) has been introduced into the teaching level evaluation to avoid the traditional research process of inductive and deductive.This study simplifies the multi-classification process of nonlinear problems using high-dimensional space mapping,reduces expected risk using the structural risk minimization theory.The research method takes advantage of the efficiency on learning small samples to achieve excellent learning outcomes and reduce the influence of the traditional evaluation methods in terms of error and subjective factors.The research result can be combined with information entropy,fuzzy math and other research methods to further enhance the accuracy of data fitting.The research has some reference value in the related areas of promoting teaching quality and improving educational evaluation.更多还原