An Optimized Mining Algorithm for Analyzing Students’ Learning Degree Based on Dynamic Data

With the rapid development of educational informatization, it has enabled education to enter the era of big data. How to extract effective information from educational big data and realize adaptive personalized learning goals have become the current research hotspot. The traditional static data only analyzes the students' learning degree based on the students' final answer, but ignores the dynamic data in the process of answering questions, such as the modification and the time it answered on the question, which makes it difficult to fully and accurately mine the correlation between the massive data, so it turns from static data mining to dynamic data mining. The paper proposes an optimized mining algorithm for analyzing students' learning degree based on dynamic data. The algorithm first uses the optimized text classification technology to match the question texts to the knowledge points automatically, so as to improves the efficiency and quality. Then, it uses the subjective weighting method combined with the expert experience to generate the learning degree matrix of students on knowledge points based on dynamic data of the students' records. Finally, the DBSCAN clustering algorithm is used to cluster the personalized learning characteristics of students according to the learning degree matrix. The experimental result shows that the algorithm can deal with massive data automatically and effectively, and analyze the students' learning degree on knowledge points comprehensively and accurately, so as to classify students and realize personalized teaching.