Scholarly articles for comparison of xg boost classifier with logistic regression algorithm to improve accuracy in student mental health prediction

Aim: The purpose of this study is to develop a reliable forecast model for student mental health. This model would offer important insights to educators and mental health professionals, helping them identify and support students who may be struggling with their mental health. By doing so, we hope to promote better mental health outcomes for students. Materials and Methods: This study assesses the predictive capability of XGBoost Classifier andLogistic Regression models for student mental health, using the "Student Mental Health" dataset obtained from Kaggle. The dataset comprises 10 pertinent columns of information. Prior to analysis, the dataset underwent several preprocessing steps including feature scaling, one-hot encoding, and outlier management, and was subsequently split into training and testing sets. The models' predictive performance was assessed through accuracy metrics and statistical analysis was conducted utilizing SPSS software. The sample sizes for both groups were calculated using clincalc.com. Results: This findings of this study indicates that machine learning algorithms are capable of accurately predicting a Mental Health of a Student. The significance value for this study is p=0.001, where is p<0.05. Hence, there is a statistically substantial dissimilarity between the two groups. Conclusions: In conclusion, findings from this investigation indicate that the Logistic Regression model has exhibited promise and efficacy in precisely predicting student mental health. Therefore, it could be advantageous to delve deeper into this approach in future studies.