Predictive model for the identification of activities of daily living (ADL) in indoor environments using classification techniques based on Machine Learning

AI-based techniques have included countless applications within the engineering field. These range from the automation of important procedures in Industry and companies, to the field of Process Control. Smart Home (SH) technology is designed to help house residents improve their daily activities and therefore enrich the quality of life while preserving their privacy. An SH system is usually equipped with a collection of software interrelated with hardware components to monitor the living space by capturing the behavior of the resident and their occupations. By doing so, the system can report risks, situations, and act on behalf of the resident to their satisfaction. This research article shows the experimentation carried out with the human activity recognition dataset, CASAS Kyoto, through preprocessing and cleaning processes of the data, showing the Vía Regression classifier as an excellent option to process this type of data with an accuracy 99.7% effective

• Autores: García-Restrepo, Johanna; Ariza-Colpas, Paola Patricia; Oñate-Bowen, Alvaro Agustín; Suarez-Brieva, Eydy del Carmen; Urina-Triana, Miguel; De-la-Hoz-Franco, Emiro; Díaz-Martínez, Jorge Luis; Butt, Shariq Aziz; Molina_Estren, Diego
• Fecha de publicación: 2021
• Institución: Universidad Simón Bolívar
• Repositorio: Repositorio Digital USB
• Idioma: eng
• OAI Identifier: oai:bonga.unisimon.edu.co:20.500.12442/8604
• Acceso en línea: https://hdl.handle.net/20.500.12442/8604; https://doi.org/10.1016/j.procs.2021.07.069; https://www.sciencedirect.com/science/article/pii/S1877050921014721?via%3Dihub
• Palabra clave: HAR; Human Activity Recognition; Machine Learning; ADL; Activity Daily Living
• Rights: openAccess
• License: Attribution-NonCommercial-NoDerivatives 4.0 Internacional