基于改进CNN算法的人力资源自动配置方法  

Human Resources Automatic Allocation Method Based on Improved CNN Algorithm

作　　者：雷江平 许涛 王志勇 LEI Jiang-ping;XU Tao;WANG Zhi-yong(State Grid Zhejiang Electric Power Co.,Ltd.,Huzhou Power Supply Company,Huzhou 313000 China)

机构地区：[1]国网浙江省电力有限公司湖州供电公司,浙江湖州313000

出　　处：《自动化技术与应用》2022年第12期142-146,共5页Techniques of Automation and Applications

摘　　要：针对电网公司人力资源自动配置失调,电网公司经济效益提升速度慢等问题,提出基于改进CNN算法的人力资源自动配置方法。通过在基础CNN模型结构内引入调节层,建立人力资源自动预测模型,利用该模型对人力资源评分矩阵实施局部特征化处理后,输出人力资源自动预测结果;依据模糊数学原理,获取员工客观评价、建立电网公司岗位素质需求矩阵和求解,实现人力资源自动配置。实验结果表明:该方法受数据稀疏度和移动步长影响较小,电网公司人力资源自动配置预测能力强;自动配置效果好,电网公司经济效益增加明显。Aiming at the maladjustment of human resources automatic allocation of power grid companies and the slow improvement of economic benefits of power grid companies, a human resources automatic allocation method based on improved CNN algorithm is proposed. By introducing the adjustment layer into the basic CNN model structure, the human resources automation prediction model is established. After the human resources scoring matrix is locally characterized by the model, the human resources automation prediction results are output;According to the principle of fuzzy mathematics, it obtains the objective evaluation of employees, establishes and solves the post quality demand matrix of power grid company, and realizes the automatic allocation of human resources. The experimental results show that this method is less affected by data sparsity and moving step size, and has strong ability to predict the automatic allocation of human resources in power grid companies;The effect of automatic configuration is good, and the economic benefit of power grid company increases significantly.

关 键 词：卷积神经网络 配置方法 模糊数学 局部特征优化 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]