Abstract:
Insomnia is a common and concerning issue among university students that can have a
significant impact on their academic performance, physical and mental health. Early
detection and treatment of insomnia is important to reduce its harmful effects. This
study uses machine learning techniques to provide a novel approach of predicting the
severity of insomnia among university students. Our research uses a dataset that was
collected from a wide range of university students and includes demographic
information, lifestyle variables and mental health indicators. In order to develop
predictive models for insomnia levels, we utilize some of the popular machine learning
algorithms, such as K-Nearest Neighbors, Support Vector Machine (SVM), Naïve
Bayes, Linear Discriminant Analysis, Stochastic Gradient Descent, Extra Trees
Classifier, AdaBoost Classifier, Ridge Classifier. All of the classifiers predict with high
accuracy and Support Vector Machine (SVM) outperformed the other models with an
excellent accuracy of 94.34%. The results show the effectiveness of machine learning
models in accurately detecting insomnia severity among university students. We can
also learn more about the factors that are strongly associated with insomnia by
conducting qualitative research. The results of our study may be utilized to develop
technology-based solutions that detect and assist students who are experiencing
sleeplessness, which will improve their academic performance and general health.
