A novel multi-scale fractional-order Bernoulli grey model and its application on primary school scale prediction

Abstract

Primary education plays an important role in the whole educational structure. The prediction of the development scale of primary schools is of great significance to the optimal allocation of educational resources by education departments. The paper introduces a novel multi-scale fractional-order Bernoulli grey model by combining a local cumulative operator with a global cumulative operator, abbreviated as MFGM. First, a nonlinear grey Bernoulli model based on conformable fractional derivative is proposed, where a hybrid accumulation is presented. Then, the MFGM parameters are obtained by an PSO optimizer, which are adaptive to data sets. Finally, the proposed model is validated and used to predict the scale of primary school scale. Experimental results show that the MFGM is superior to the other grey models and machine learning models in the fitting and testing primary school scale. The prediction results contribute to the decision-making of education sectors.

Keywords:

• Fractional order
• Grey model
• Prediction
• Primary school scale

Disclosure statement

No potential conflict of interest were reported by the author(s).

Data availability statement

The data that support the findings of this study will be available in http://www.stats.gov.cn.

Additional information

Funding

The work in this paper was supported by the National Natural Science Foundation of China (Grant No. 62007020), 2023 Xuzhou Soft Science Project” Research on the Introduction and Training Mechanism of Young Science and Technology Talents in Xuzhou” (No. KC23099), Doctoral Research Foundation of Jiangsu Normal University (No. 21XSRX005), China Postdoctoral Science Foundation (No. 2022M711883), and Taishan Scholar Project of Shandong Province (No. tsqn202211130).