Article Details

Abstract

Taking the lack of economic theory support in traditional Machine Learning classification models as a starting point, this paper proposes a Value-Driven Classification (VDC) model based on Support Vector Machine (SVM). Through deep combination with Marxist value theory, the model innovatively analogizes data samples to commodities and labor power, constructs a feature value quantification system and a surplus value fluctuation calculation model, and uses kernel functions to process complex features. Experimental results show that the model enhances theoretical interpretability while maintaining classification performance comparable to traditional models. This result not only expands the theoretical basis of Machine Learning models but also provides new methods for quantitative economic research, and has application potential in fields such as economic empirical analysis and market prediction.

Keywords

classification model; value; VDC model; Python language; SVM model; kernel function

References

1. [1] Jian Xinhua Rethinking the Labor Value Theory and Surplus Value Theory [J]. Journal of Hebei University of Economics and Business, 2024, 45 (6): 1-13
2. [2] Zhang Yi, Sun Peipei, Wang Xianjia Evaluation of Water Resources Carrying Capacity in Hubei Province Based on Entropy Mutation Series Method [J]. Hydroelectric Energy Science, 2025, 43 (2): 49-53
3. [3] Chen Xi Research on the Evaluation Model of Volleyball Training Effectiveness Based on Information Gain and Random Forest Algorithm [J]. Journal of Kashgar University, 2024, 45 (6): 79-84
4. [4] Wang Yan, Liu Banteng Research on the Deployment Model of Wireless Sensor Network Nodes Based on Gradient Descent Algorithm [J]. Journal of Sensing Technology, 2024, 37 (11): 1989-1993
5. [5] Song Zhiqiang, Jie Lidong, Wang Mingyang, etc Optimization of Support Vector Machine for Power Communication Network Fault Diagnosis Based on Ant Lion Algorithm [J]. Automation Technology and Application, 2025, 44 (8): 29-32
6. [6] Zhao Jinghua, Wen Long, Wang Shoufeng, etc Vehicle speed prediction based on improved combination kernel function Gaussian process regression [J]. Journal of Jilin University: Science Edition, 2025, 63 (2): 454-464
7. [7] Kong Yinghui, the scenery is beautiful Research on Classification Methods Based on Confusion Matrix and Ensemble Learning [J]. Computer Engineering and Science, 2012, 34 (6): 111-117
8. [8] Yao Hui, Yin Shangxian, Xu Wei, etc Risk assessment of floor water inrush using weighted rank sum ratio method based on combination weighting [J]. Coalfield Geology and Exploration, 2022, 50 (6): 132-137
9. [9] Xie Binyao A method for identifying tobacco maturity based on image processing [D]. Chongqing: Southwest University, 2020
10. [10] Gao Rui Research on the Application of Confusion Matrix in Business Decision Making [J]. Contemporary Economics, 2021 (4): 110-113
11. [11] JOLAYEMI J K，FLOTT P. Handbook of Credit Scoring [J].The Journal of Developing Areas，2003，36（2）：149-152.
12. [12] Cao Min, Yin Shangxian, Li Shuqian, etc Risk assessment of water inrush from mining floor under pressure based on confusion matrix [J]. Coal Science and Technology, 2025, 53 (9): 407-417
13. [13] Lu Wei Machine Learning Formula Deduction and Code Implementation [M]. Beijing: People's Posts and Telecommunications Press, 2022
14. [14] Bai Zhihong, Hu Yonggui, Zhang Yang, etc Method for evaluating the uniformity of observation systems based on heat map analysis [C]//Oil and Gas Geophysics Professional Committee of the Chinese Geophysical Society Proceedings of the 6th Annual Conference on Oil and Gas Geophysics Beijing: China National Knowledge Infrastructure, 2024: 252-255
15. [15] Jiang Man Product type bipolar fuzzy point ideals of Boolean algebra [J]. Journal of Capital Normal University: Natural Science Edition, 2022, 43 (4): 15-18
16. [16] Cao Hongbao, Peng Jinrui, Hui Jiaojiao Research on the Current Status of Intangible Cultural Heritage Communication Based on Factor Analysis and BP Neural Network: A Case Study of Shaanxi Province [J]. Journal of Capital Normal University: Natural Science Edition, 2023, 44 (5): 24-29
17. [17] Hui Jiaojiao, Cao Hongbao, Zhang Zhe Research on the Communication of Intangible Cultural Heritage Based on Multiple Logit Regression [J]. Journal of Weinan Normal University, 2023, 38 (2): 85-93

About the authors

Cao Hongbao (2001-), male, Han ethnicity, from Yulin, Shaanxi, engineer, Bachelor of Science, research interests include statistics and machine learning.