Accurate prediction of college students' information anxiety based on optimized random forest and category boosting fusion model

The paper aims to construct an efficient predictive model to accurately predict information anxiety among college students and provides a scientific basis for mental health interventions. Firstly, the random forest algorithm is used to preprocess the relevant data and select the best features, and then the prediction model is built based on the CatBoost algorithm. To solve the problem of the imbalance of the dataset, a few class oversampling techniques are introduced, and the parameters of the model are optimized. The results showed that the model integrating random forest and category boosting algorithm had the lowest mean absolute error and root mean squared error, which were 0.125 and 0.142. Meanwhile, the area value under the receiver operating characteristic curve was close to 1, demonstrating excellent classification performance. The fusion model achieved an interpretable variance value of 0.923, an R2 value of 0.918, and a Logloss of 0.082, confirming its advantages in prediction accuracy and model fit. The fusion model had higher accuracy and stability in predicting information anxiety psychology among college students. This model can provide effective decision support for mental health educators, assist in early identification and intervention of information anxiety, and promote the mental health development of college students.