Construction of Demand Forecasting Model of Human Resources Professional Structure Based on Deep Learning

In the aspect of HR (human resources) management, DL (Deep learning) has been competent for HR management, such as recruitment, salary calculation and performance management, and will soon occupy a place in HR planning, training and development, and employee service, and will realize the high intelligence of HR service in the future. Therefore, this paper puts forward a demand forecasting model of HR professional structure based on DL. Firstly, BiLSTM is used to encode the text input, and then the transformed features are captured by attention mechanism and mixed convolution operation, which can capture the key information of the text. Realize the scientific, reasonable and accurate prediction of the future HR professional structure requirements. The experimental results show that the DL model proposed in this paper is 0.015 higher in accuracy, 0.094 higher in F1 value and 0.088 higher in AUC than only using artificial neural network model to extract features for text matching. Forecasting the demand of HR professional structure based on DL model can provide data support for enterprises to study the change and trend of talent demand in the field, and provide decision-making basis for higher matching talent training.