School of Business, Anhui University, Anhui, Hefei 230601, China.
Comput Intell Neurosci. 2022 May 19;2022:1418020. doi: 10.1155/2022/1418020. eCollection 2022.
This article uses deep neural network technology and combines digital HRM knowledge to research human-job matching systematically. Through intelligent digital means such as 5G communication, cloud computing, big data, neural network, and user portrait, this article proposes the design of the corresponding digital transformation strategy of HRM. This article further puts forward the guaranteed measures in enhancing HRM thinking and establishing HRM culture to ensure the smooth implementation of the digital transformation strategy of the HRM. This system uses charts for data visualization and flask framework for background construction, and the data is stored through CSV files, My SQL, and configuration files. The system is based on a deep learning algorithm for job applicant matching, intelligent recommendation of jobs for job seekers, and more real help for job applicants to apply for jobs. The job intelligent recommendation algorithm partly adopts bidirectional long and short-term memory neural network (Bi-LSTM) and the word-level human post-matching neural network APJFNN built by the attention mechanism. By embedding the text representation of job demand information into the representation vector of public space, a joint embedded convolutional neural network (JE-CNN) for post matching analysis is designed and implemented. The quantitative analysis method analyzes the degree of matching with the job.
本文运用深度神经网络技术，融合数字人力资源管理知识，系统地对人岗匹配进行研究。通过 5G 通信、云计算、大数据、神经网络、用户画像等智能数字手段，提出了人力资源管理数字化转型策略的相应设计。本文进一步提出了增强人力资源管理思维和建立人力资源管理文化的保障措施，以确保人力资源管理数字化转型策略的顺利实施。该系统使用图表进行数据可视化，并使用 flask 框架进行后台构建，通过 CSV 文件、My SQL 和配置文件进行数据存储。该系统基于求职者匹配的深度学习算法、求职者工作的智能推荐等，为求职者申请工作提供更真实的帮助。工作智能推荐算法部分采用双向长短期记忆神经网络（Bi-LSTM）和基于注意力机制的词级人工岗位匹配神经网络 APJFNN。通过将工作需求信息的文本表示嵌入公共空间的表示向量中，设计并实现了用于岗位匹配分析的联合嵌入卷积神经网络（JE-CNN）。采用定量分析方法对与工作的匹配程度进行分析。
