Study Of Shandong Undergraduate Students Subjective Well-being Via Machine Learning Approaches

Aim:This work aims to analyze undergraduates’ subjective well-being by machine learning approaches and to discover major factors that have impact on undergraduates’ subjective well-being.Using those factors,we can predict undergraduates’ subjective well-being.Those factors may also server as the reference direction for the development of psychological counseling in colleges and universities.Methods:The design and curation of the questionnaire involved network questionnaire production,which includes 10 scales related to happiness,personal information and other questions.After the completion of the questionnaire,we the happiness team cooperated with Jining medical college of Shandong province to collect questionnaire data of college students.After sorting and cleaning the data,the computer language python is used as a programming tool,and the feature selection algorithm in machine learning is used to screen the factors that have a higher impact on undergraduates’subjective well-being.The gradient ascending classification classification tree algorithm in machine learning is adopted to build the prediction model.Results:301 factors that may affect undergraduates’ subjective well-being were included in the questionnaire,and 10518 questionnaires were collected.Top 20 factors that have a statistically significant impact on undergraduates’ subjective well-being were included in the final model.The prediction model established by gradient ascending classification classification tree algorithm.The area under curve(AUC)is 0.96.And the model can achieve 90%accuracy,92%sensitivity and 89%specificity.Conclusion:This study adopts machine learning method in computer science to analyze and predict undergraduates’ subjective well-being.This study makes statistics on the subjective well-being of college students and analyzes the influencing factors.In addition,analyzing high influence factors may provide advices for the mental health treatment and prevention of college students.