Akademik Veri Yönetim Sistemi
International Conference on Computing, Intelligence and Data Analytics, ICCIDA 2022, Kocaeli, Türkiye, 16 - 17 Eylül 2022, cilt.643 LNNS, ss.358-369
Sleep is not linear and contains different stages. The quality of sleep and time spent in each sleep stage directly affect a person’s daily life. Sleep quality can be defined as how well an individual rested during the time he or she slept. In order to measure the sleep quality, sleep monitoring is used and the main focus of most of the works about sleep monitoring in the literature is the effects of environmental factors on sleep. Some of the previous works consider the effects of physical exercise or addiction on sleep quality. The works that are focusing on the effects of diseases on sleep quality usually examine a single disease. Unlike previous works in the literature, the main goal of our work is to examine different factors - such as multiple medications and different diseases - altogether and compare these factors to each other. In this paper, we analyzed patient data to reach our goal. A variation of the Cross Industry Standard Process for Data Mining (CRISP-DM) methodology is used to conduct analyses. K-means clustering algorithm is used for clustering the data. Random forest classifier is used in order to determine feature importance. Our analyses of patients showed that diseases they have, medications they use, and their age and gender have the most impact on sleep quality. Diseases that change breathing patterns and depression affect the sleep quality mostly. Different types of the antidepressants such as SSRI and trazodone have an important impact. Also, epilepsy is one of the factors that affect sleep quality.