Abstract
Enterprise employees are confronted with the dual challenges of social competition and work pressure, and mental health problems are becoming increasingly prominent. Financial staff play a very important role in an enterprise, and their mental health issues can have negative impacts on the operational efficiency and productivity of the enterprise. However, current methods for analyzing mental health suffer from high costs, low accuracy, and low efficiency. In response to this, a financial staff mental health analysis model that combines biosensing technology and decision tree algorithm is proposed. This model is applied to enterprise management work. The study first takes the biomechanical feature of pulse waves as the analysis basis, collects pulse wave signals using photoelectric sensors, and analyzes the variability of pulse rate. Afterwards, the variability features are input into a deep forest model integrated based on decision tree algorithm for psychological state recognition. The study selected 100 employees as the research subjects. Through a 3-month follow-up monitoring period, data was collected and the validity of the model was verified. The F1 score, recognition accuracy, recognition time, and recall rate were 0.94, 95.28%, 2.31 seconds, and 93.84%, respectively. Compared with other models, its performance was significantly better. In addition, in practical application experiments, the psychological health analysis accuracy of the proposed model reached 95.35%. In practical applications, the turnover rate and performance of the financial staff in the corresponding enterprises decreased and increased by 8.09% and 10.20%, respectively. The proposed model can further assist companies in monitoring the mental health issues of the financial staff and enhance their management capabilities.
Get full access to this article
View all access options for this article.