9th International Conference on ICT for Sustainable Development

Authors - Revathy P, Manju Bhargavi N, Gunasekar S, Lohit A
Abstract - Sleep disturbances represent a significant health concern, with widespread implications for overall well-being. The study involves exploratory analysis to understand the relationships and associations between sleep disturbances and various chronic disorders, mental disorders, and demographics of individuals. Leveraging a rich dataset encompassing respondent sequence numbers and diverse sleep-related and health-related variables, this study employs to extract essential variables for further analysis. The predictive analysis involved the use of robust visualization and statistical tools to better understand the major distributing agents of sleep disturbances in the participants of the NHANES questionnaire. The second phase of this project focuses on fine-tuning the predictive model developed in the initial phase to enhance its accuracy and reliability. Building upon the insights gleaned from exploratory analysis and initial modelling, this phase employs advanced techniques to refine the predictive capabilities of the model. Specifically, the methodology involves the implementation of a voting regressor approach, wherein a diverse set of regression models are built and combined to make the final prediction. By leveraging the collective wisdom of multiple models, the voting regressor technique aims to mitigate individual model biases and uncertainties, ultimately yielding a more robust prediction of sleep disturbance likelihood. Through rigorous evaluation and validation, this phase seeks to optimize the performance of the predictive model, ensuring its effectiveness in real-world scenarios. The outcomes of this phase contribute to the overarching goal of harnessing machine learning to better understand and predict sleep disturbances, thereby facilitating proactive interventions and improving overall health outcomes.