Exploration of general education work in universities based on the feature extraction algorithm

Abundant general education resources meet the needs of learners to independently choose learning content. However, the resources themselves and their platforms have problems such as uneven content quality, lack of high-quality educational resources, homogenization of resources, and single method of resource selection. It is difficult for learners to quickly and efficiently obtain information related to their own needs from the massive teaching resources and valuable resources. To allow learners to efficiently obtain personalized general education resources, when selecting resources, the study adopts deep learning methods to accurately identify the knowledge points in general education resources, by constructing a relationship between the knowledge points in the teaching resources and the needs of learners. The feature vector is used as the input of the support vector machine, and the support vector machine decides whether to recommend the content of general education to learners. The actual click rate of the learner's push results and the satisfaction of the learner's feedback are used as the performance evaluation index of the method. This push method pays attention to the combination of learners' interest needs and the characteristics of general education courses, which can better meet the requirements of learners and improve learning efficiency, and has greater application potential.