Chemical Experiment Aided Design System Based on Data Mining Algorithms

Experimental teaching refers to the instructional stage of combining theoretical knowledge with practical activities, abstraction with concreteness, and direct and indirect. This teaching mode is more comprehensive, practical, direct and innovative than pure theory teaching. In this article, a instructional resource recommendation model of networked chemistry experiment instructional platform based on improved collaborative filtering (CF) algorithm is proposed. The user preferences are modeled and analyzed, and the obtained user model vectors are clustered, so as to realize individuation recommendation of chemistry experiment instructional resources driven by big data and provide support for networked chemistry teaching in universities. The results show that the recommended model of chemical experiment instructional resources in this article has lower error and higher recommendation accuracy. Compared with the traditional CF algorithm, the average absolute error (MAE) is reduced by 19.78% and the recall is increased by 24.56%. The comprehensive experimental results show that the recommended model of chemical experiment instructional resources in this article has high accuracy and low error performance, and can be applied to the network construction of chemical experiment teaching to provide technical support for the construction of chemical experiment aided design system.