Building a Korean Language Teaching Resource Library Based on Python Crawler

Abstract

Nowadays, with the popularization of online education how to efficiently and accurately obtain the required educational resources from web pages is one of the key concerns in the teaching process. Based on the principle of maximum distance, the study improves the K-mean and establishes a learner group feature model based on the DM-K-mean clustering algorithm. In addition, for the problem that the importance of learning resources to learners changes with time, this paper integrates the time information into the neural collaborative filtering algorithm through the clustering classification algorithm. It proposes a deep learning-based recommendation algorithm for Korean language teaching resources. Python crawler technology is used to obtain relevant experimental data from the online teaching platform to verify the performance of the proposed model, so as to construct the Korean language teaching resource base. The learner group characteristic model classifies the sample students into three categories: excellent (0.489), good (0.307), and average (0.204) learning situations. The HR and NDCG values of the Improved NeuMF resource recommendation model have been improved by 3.6% and 2.2% compared to the NeuCF model, respectively, and the performance is optimal under various factors. The proposed system for teaching Korean language resources in this paper can recommend resources based on learner profiles to help learners access teaching resources and improve efficiency.