Using Artificial Intelligence to Predict Teenagers' Stress Levels

Abstract

Research on adolescent stress helps accurately clarify the physical and mental development rules of teenagers and facilitate their healthy and smooth growth. However, current relevant studies are limited by overly single measurement tools and an overreliance on self-report data. The data of this study is taken from the author's research data set in Kaggle, which contains multi-dimensional indicators and analysis requirements of modeling and experimental results, and using the Logistic Regression model, the Random Forest, the Gradient Boosting artificial intelligence model, three different models were used for experimental analysis and the accuracy of the three models was compared, so that artificial intelligence was used to predict the stress level of adolescents and it is found that The Random Forest model achieves the optimal prediction performance. The experiment not only verifies the effectiveness of artificial intelligence in predicting the stress level of adolescents, but also reflects the trend of the current stress level of adolescents, which is of great significance to the future development of adolescents.