Water Resources Big Data Classification Based on Multi-Objective Optimization for Mining Area
Keywords:
Support Vector Machine, Big Data Classification, Soft Computing, Mining Area, Particle Swarm Optimization.Abstract
In order to solve the uncertainty of support vector machine kernel function parameters and solve the optimal selection of kernel parameters in the classification algorithm of big data of mining area water resources, a mining area water big data classification algorithm based on PSO-SVM hybrid optimization is proposed. This algorithm solves the existence of inseparable regions and error accumulation in the support vector machine multi-classification method. Based on the analysis of basic particle swarm optimization algorithm and SVm algorithm working principle, the advantages of PSO and SVM algorithm are mixed, and the convergence speed is moderately improved to make it have the ability of self-adaptation, and the fine search is performed in the final stage. Expand the width and depth of parameter search to meet the characteristics of diversification and concentration. The results show that the hybrid soft calculation method proposed in this paper can improve the accuracy of classification and prediction, and classification accuracy and classification time are significantly improved, and it is an effective multi classification algorithm.
Downloads
Metrics
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
References
Liu S S, Zhang H, Mao Z, et al. (2014): Target detection method based on HRM extracting and SVM, Foreign Electronic Measurement Technology, 33(10), 38-41.