引力搜索优化ELM的企业财务危机预警方法

Extreme learning machine based on gravitational search algorithm optimization for enterprise crisis prediction

针对企业财务危机进行了研究,提出一种基于引力搜索算法优化核极限学习机(KELM)的并行模型PHGSA-KELM。模型考虑了特征选择机制和参数优化两者对KELM模型起着同等重要的作用,提出改进的引力搜索算法(HGSA)同步实现特征选择机制和KELM参数优化,同时设计的线性加权多目标函数综合考虑了分类精度和特征子集数量,改善了算法的分类性能,并且基于多核平台的多线程并行方式进一步提高了算法的计算效率。通过真实数据集的实验结果表明,提出的模型不仅获得了较少的特征子集个数,找出了与企业财务危机紧密相关的特征,得到了很高的分类准确率,并且计算效率也得到极大提高,是一种有效的企业财务危机预警模型。

To improve the precision rate of enterprise financial crisis, this paper developed a novel gravitational search algorithm based kernel extreme learning machine （KELM） parallel model PHGSA-KELM. The model used to solve the problem of feature selection and parameter optimization separately. This paper applied an improved gravitational search algorithm （HGSA） to conduct feature selection and parameter optimization simultaneously. It used a linear-weighted multi-objective function to improve the accuracy of the algorithm, taking into account the average accuracy rate and the subset of feature selection. Moreover, HGSA-KELM model implemented in parallel on multi-core processor, which used OpenMP to speed up the search and optimization process. Real datasets were used to verify this method and the results of simulation, which compared to some similar algorithms. It indicates that this model is better than several other algorithms, and achieves small subset of features. It selects the most related features of enterprise financial crisis, and improves the accuracy and efficiency. Simulation experiments show that the proposed model is effective and efficient.