| 摘要: |
| 企业二次创业金融数据的优化分类能提高数据的统计分析能力，提出基于逻辑回归的企业二次创业金融数据分类方法，采用自适应无监督学习的方法进行数据统计的融合处理，构建数据分布的不规则空间聚类模型，采用相空间结构重组方法进行数据的模糊特征重构，提取企业二次创业金融数据的关联规则特征量，采用逻辑回归分析方法进行数据分类的融合聚类处理，结合模糊C均值聚类方法，实现对数据分类的自适应寻优和收敛性控制，实现数据分类优化。仿真结果表明：采用该方法进行企业二次创业金融数据分类的准确性较高，收敛性较好，特征聚类能力较强。 |
| 关键词: 逻辑回归 企业二次创业 金融数据 分类 |
| DOI： |
| 分类号: |
| 基金项目: |
|
| Research on Classification Method of Enterprise Secondary Entrepreneurship Financial Data Based on Logical Regression |
|
LAI Hong-qing
|
|
Department of Business and Trade, School of Business Administration, Foshan Polytechnic CollegeGuangdong Foshan 528200, China
|
| Abstract: |
| The optimal classification of enterprise secondary venture finance data can improve the statistical analysis ability of the data. The classification method of enterprise secondary venture finance data based on logical regression is put forward. The self-adaptive unsupervised learning method is used to carry on the statistical fusion processing of enterprise secondary venture finance data, and the irregular spatial clustering model of enterprise secondary venture finance data distribution is constructed. The fuzzy feature reconstruction of enterprise secondary venture finance data is carried out by using phase spatial structure reorganization method, the self-association rule feature quantity of enterprise secondary venture finance data is extracted, and the fusion clustering processing of enterprise secondary venture finance data classification is carried out by using logical regression analysis method. Combined with fuzzy C-means clustering method, the adaptive optimization and convergence control of enterprise secondary venture finance data classification are realized. To realize the classification and optimization of the secondary entrepreneurial financial data of enterprises, the simulation results show that the method has the advantages of high accuracy, good convergence and strong feature clustering ability. |
| Key words: logical regression enterprise secondary entrepreneurship financial data classification |
