<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>In this paper, after dissecting the neural network model, the initial weights and thresholds of the BP neural network are optimized through selection, crossover, mutation and other operations by using the global optimization-seeking ability of the genetic algorithm. The model for predicting employee mental health risks is initially constructed by selecting structural design, structural parameters, and genetic operators. The feasibility of the model in mental health risk prediction was explored based on the indicators of training time, model error and prediction accuracy, and the prediction model was utilized to predict the mental health risk of the employees in Company A and the coping plan for the employees’ mental health risk was established. The results show that the accuracy of the neural network model is 90% and 85%, respectively, during the training and testing processes. The sensitivity and specificity of the training set are 90.00% and 75.06%, respectively, the Yoden index is 0.78, and the Kappa coefficient is 0.69, and the sensitivity and specificity of the test set are 92.00% and 78.05%, which is a good performance, based on which the study is able to predict the risk of mental health of the employees and to guarantee the mental health of the employees in real-time.</p>
</abstract>

In this paper, after dissecting the neural network model, the initial weights and thresholds of the BP neural network are optimized through selection, crossover, mutation and other operations by using the global optimization-seeking ability of the genetic algorithm. The model for predicting employee mental health risks is initially constructed by selecting structural design, structural parameters, and genetic operators. The feasibility of the model in mental health risk prediction was explored based on the indicators of training time, model error and prediction accuracy, and the prediction model was utilized to predict the mental health risk of the employees in Company A and the coping plan for the employees’ mental health risk was established. The results show that the accuracy of the neural network model is 90% and 85%, respectively, during the training and testing processes. The sensitivity and specificity of the training set are 90.00% and 75.06%, respectively, the Yoden index is 0.78, and the Kappa coefficient is 0.69, and the sensitivity and specificity of the test set are 92.00% and 78.05%, which is a good performance, based on which the study is able to predict the risk of mental health of the employees and to guarantee the mental health of the employees in real-time.

Employee mental health risk prediction and coping management based on neural network

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

In this paper, after dissecting the neural network model, the initial weights and thresholds of the BP neural network are optimized through selection,...

Employee mental health risk prediction and coping management based on neural network

Publicado en línea: 23 dic 2023

Páginas: -

Recibido: 16 feb 2023

Aceptado: 03 jul 2023

DOI: https://doi.org/10.2478/amns.2023.2.01601

Palabras clave
Neural network, Genetic algorithm, Global optimization, Prediction accuracy, Mental health risk

© 2023 Lijun Chang, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Employee mental health risk prediction and coping management based on neural network

Publicado en línea: 23 dic 2023

Páginas: -

Recibido: 16 feb 2023

Aceptado: 03 jul 2023

DOI: https://doi.org/10.2478/amns.2023.2.01601

Palabras claveNeural network, Genetic algorithm, Global optimization, Prediction accuracy, Mental health risk

© 2023 Lijun Chang, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Palabras clave
Neural network, Genetic algorithm, Global optimization, Prediction accuracy, Mental health risk