Construction of university big data mining and analysis model based on FP-growth algorithm

Abstract

To deeply explore the potential correlation between in-class grades and extracurricular training plans, a big data mining and analysis model for colleges and universities based on the improved FP-growth algorithm is studied and proposed. Based on the traditional FP-growth algorithm, this model integrates the C4.5 partitioning strategy and optimizes the FP-growth tree structure, significantly improving the mining efficiency and accuracy of big data in colleges and universities. Compared with the traditional FP-growth association rule algorithm, the running time of the research model is only 9 minutes, which is 12 minutes shorter. The simulation results show a strong correlation between the in-class grades and the grades of social practice and campus cultural activities. The confidence levels exceed 80%, the accuracy rate reaches 92.32%, and the loss value is only 0.18. The accuracy rate of the research model increases by 17.97% compared with the traditional model. From this, the model proposed by the research has excellent data mining and data analysis capabilities, and can provide a new suggestion and direction for student management in the field of education.