Application of data visualization technology in human resource management and employee resignation prediction

Accurately predicting the likelihood of employee turnover is crucial for companies to reduce key talent loss and improve employee satisfaction and performance. The reasons for resignation are studied and analyzed, and a prediction model based on an improved decision tree and cross-validation integration is constructed. Visual analysis indicated that companies needed to strengthen retention measures, increase salaries, improve working environments, and meet employee demands to reduce turnover. The model performance test showed an error rate as low as 3.46%, an accuracy rate of 96.54%, a precision rate of 99.12%, a recall rate of 99.23%, an F1 value of 99.17%, and an AUC of 0.914. The research model was significantly better than Naive Bayes, logistic regression, and gradient boosting tree models in core indicators such as error rate (maximum reduction of 14.89%) and accuracy. This model has high accuracy, credibility, and predictive ability, which can be effectively applied to human resource management.