Author(s)

Harriet Kim, Haijun Bai^*, Zandong Sun

Affiliation(s)

China University of Petroleum, Dongying 257061, Shandong, China.

Corresponding Author

Haijun Bai

ABSTRACT

Support vector machine improves the generalization ability through the principle of structural risk minimization. Now it is mostly used to solve the classification and regression problems of small samples, but when it is used for prediction, a single model has certain limitations. This paper proposes a prediction model that combines the improved multidimensional gray model and support vector machine to realize the complementary advantages of different models, avoid the limitations of a single model, and increase the stability of the model. Experimental simulation results show that the forecasting effect of the proposed combined forecasting model is significantly better than that of support vector machine and the model based on innovation priority accumulation method, and the forecasting accuracy is higher than that of the single forecasting model.

KEYWORDS

Forecasting model; Multidimensional gray model; Combination of support vector machines; Forecasting model.

CITE THIS PAPER

Harriet Kim, Haijun Bai, Zandong Sun. Combined prediction of improved multidimensional gray model and support vector machine. World Journal of Mathematics and Physics. 2023, 1(1): 1-7.

REFERENCES

[1] Wu Huaan, Zeng Bo, Peng You, Zhou Meng. Prediction of urban population density based on multidimensional gray system model. Statistics and Information Forum, 2018, 33(08): 60-67.

[2] Luo Yiyong, Zhang Hao, Zhang Liting. Research on Multidimensional Gray Deformation Prediction Model Based on Genetic Support Vector Machine. Journal of Zhejiang University of Technology, 2010, 3 8 (01): 79-83.

[3]Awad M, Khanna R. Efficient learning Machines. Apress, 2015: 203-224. 

[4]Zhang Haoran, Han Zhengzhi, Li Changgang. Support Vector Machines. Computer Science, 2002, 29(12): 135-137.

[5] Meyer D, Leisch F, Hornik K. Benchmarking Support vector Machines. Austria：Vienna University of Economics and Business Administration, 2002.

[6] Yang Guangxi. Application of combined forecasting method in economic forecasting. Statistics and Forecasting, 1998(6): 17-20.

[7] Tang Xiaowo. Optimal combination forecasting method and its application. Mathematical Statistics and Management, 1992(1): 31- 35.

[8] Chen Huayou. Combination forecasting method validity theory and its application. Beijing: Science Press, 2008: 1-7.

[9] Zeng Bo, Liu Sifeng, Fang Zhigeng. Gray combination forecasting model and its application. Chinese Management Science, 2009, 17(5): 150-155.

[10] Yuan Quan, Zeng Xiangyan. The GM(0, N) model based on innovation priority accumulation method and its application. Statistics and Decision Making, 2018, 34(12): 79-81.  

[11] Yuan Quan, Zeng Xiangyan. The GM (1, N) model based on innovation priority accumulation method and its application. Journal of Guilin University of Electronic Technology, 2017, 37(04): 332-336.

[12] Gu Yaxiang, Ding Shifei. Research progress of support vector machine. Computer Science, 2011, 38(2): 14-17.

[13] Xu Jianhua, Zhang Xue Work, Li Yanda. The new development of support vector machine. Control and Decision Making, 2004, 19(5): 481-484. 

[14] Sun Deshan. Research on Support Vector Machine Classification and Regression Method. Changsha: Central South University, 2004.

[15] Yang Junyan, Zhang Youyun, Zhu Yongsheng. ε- insensitive loss function support vector machine classification performance research. Journal of Xi'an Jiaotong University, 2007, 41(11): 1315-1320.

[16] Feng Guohe. SVM classification kernel function and parameter selection comparison. Computer Engineering and Application, 2011, 47(3): 123-124.