Volleyball teaching is one of the traditional contents of physical education in our country. It plays an extremely important role in improving students' volleyball skills, promoting their physical and mental health, and improving their character training. However, due to the influence of volleyball's own characteristics and the restriction of test-oriented education concepts, the development of volleyball teaching has been seriously lagging. It not only fails to effectively adapt to the social development situation, but also severely weakens the students' athletic ability, making volleyball learn from the students. The interest in volleyball has gradually decreased, and the audience who watched the volleyball game has also been reduced. In addition, the degree of attention has also been declining, which ultimately makes volleyball teaching a more embarrassing and marginalized situation. This article uses the method of literature review to illustrate the innovative significance of the volleyball teaching evaluation system, and discuss its related innovative development methods. Its purpose is to correct the shortcomings of the traditional teaching evaluation system, improve the effectiveness of volleyball teaching evaluation, and stimulate students to learn volleyball. Interest also builds self-confidence in learning, and provides necessary reference and reference. It is well known that selecting all kernel functions through a model that is not sparse will generate a lot of messy and unordered information and be sensitive to noise. In order to solve the above problems, this article will propose a very formal sequential learning algorithm based on the AdaBoost framework. When the basic classifier is selected iteratively, the proportion of the kernel function will be constrained by the elastic net type normalization, which is the mixed L norm and L. The norm constraint is to construct a basic classifier with the best combination of multiple basic cores, and in addition to receiving them into a powerful classifier.

排球教学是我国体育教学的传统内容之一。它对提高学生排球技术、促进身心健康、提高品格修养具有极其重要的作用。但是，由于排球自身特点的影响和应试教育理念的制约，排球教学的发展一直严重滞后。不仅不能有效适应社会发展形势，而且严重削弱了学生的运动能力，使排球向学生学习。人们对排球的兴趣逐渐降低，观看排球比赛的观众也减少了。此外，关注度也一直在下降，这最终使排球教学变得更加尴尬和边缘化。本文采用文献综述的方法来说明排球教学评价体系的创新意义，并探讨其相关的创新发展方法。其目的是为了纠正传统教学评价体系的不足，提高排球教学评价的有效性，激发学生学习排球的兴趣。兴趣也可以建立学习的自信心，并提供必要的参考和借鉴。众所周知，通过一个不稀疏的模型选择所有核函数会产生大量杂乱无序的信息，并且对噪声很敏感。为了解决以上问题，本文将提出一种基于 AdaBoost 框架的非常正式的顺序学习算法。迭代选择基本分类器时，核函数的比例会受到弹性网型归一化的约束，即L范数和L的混合。范数约束是构造多个基本核的最佳组合的基本分类器，并且除了将它们接收到一个强大的分类器中。