Predicting Sleep Disorders: Leveraging Sleep Health and Lifestyle Data with Dipper Throated Optimization Algorithm for Feature Selection and Logistic Regression for Classification

El-kenawy, El-Sayed M.; Ibrahim, Abdelhameed; Abdelhamid, Abdelaziz A.; Khodadadi, Nima; Abualigah, Laith; Eid, Marwa M.

doi:10.21608/cjmss.2024.290167.1053

Predicting Sleep Disorders: Leveraging Sleep Health and Lifestyle Data with Dipper Throated Optimization Algorithm for Feature Selection and Logistic Regression for Classification

Document Type : Original Article

Authors

¹ Department of Communications and Electronics, Delta Higher Institute of Engineering and Technology Mansoura, Egypt

² Computer Engineering and Control Systems Department, Faculty of Engineering, Mansoura University, Mansoura 35516, Egypt

³ Department of Computer Science, College of Computing and Information Technology, Shaqra University, 11961, Shaqra, Saudi Arabi

⁴ Department of Computer Science, Faculty of Computer and Information Sciences, Ain Shams University, Cairo 11566, Egypt

⁵ Department of Civil and Architectural Engineering, University of Miami, Coral Gables, FL, USA

⁶ Hourani Center for Applied Scientific Research, Al-Ahliyya Amman University, Amman 19328, Jordan

⁷ Computer Science Department, Al al-Bayt University, Mafraq 25113, Jordan

⁸ MEU Research Unit, Middle East University, Amman 11831, Jordan.

⁹ Applied science research center, Applied science private university, Amman 11931, Jordan

¹⁰ Faculty of Artiﬁcial Intelligence, Delta University for Science and Technology, Mansoura 11152, Egypt

10.21608/cjmss.2024.290167.1053

Abstract

This paper is a thorough examination of the modeling of sleep disorders based on machine learning that is applied to the sleep-health-and-lifestyle data. The use of the Dipper Throated Optimization Algorithm for feature selection and Logistic Regression for classification is the basis of the study that explores the effectiveness of predictive models in identifying sleep disorders based on varied sleep metrics and lifestyle factors. The binary Dipper Throated Optimization Algorithm was the most successful with the lowest Average error of 0.71933 uses feature selection as the most effective method, which proves that it is successful the method of choosing the relevant features for predictive modeling. Moreover, Logistic Regression proved to be very efficient in classification; it got an Accuracy of 0.95. The results of these studies support the idea of the personalized treatment and earlier detection of sleep disorders; this, in turn, will be of great help to the progress in sleep health research and healthcare practice.

Keywords