Applying Machine Learning Algorithms for the Classification of Sleep Disorders

Abstract

Sleep issues, particularly sleep apnea, can cause people to get ill and that is why it is important to have the correct diagnosis. Instead, sleep experts have complex and time-intensive methods of sorting out the various sleep stages manually. This paper introduces a ML classification algorithm based on the publicly available Sleep Disorder Data, comprising 400 records and 13 characteristics. A number of profound and technique-based ML models are discussed, and their effectiveness in accurately detecting sleep disorders is evaluated. Some of the key aspects in the dataset that can aid in finding patterns are lifestyle factors and sleep health indicators. These tendencies may indicate that a person has a sleeping problem. The analysis of the models showed that bagged models (the Voting Classifier with RF and DT) had the best performance. Accuracy, precision, recall, and F1- score of the algorithm are all 0.973 which implies that the algorithm is useful in classifying the sleep disorders and is reliable. These results suggest that the suggested ML methods offer a chance to create sleep problem diagnoses that are smarter, quicker, and more accurate. This would enhance decision-making and health of the doctors and patients.