Seflek I.2024-12-102024-12-102024979-833153149-2https://doi.org/10.1109/IDAP64064.2024.10711019https://hdl.handle.net/20.500.13091/96718th International Artificial Intelligence and Data Processing Symposium, IDAP 2024 -- 21 September 2024 through 22 September 2024 -- Malatya -- 203423Fall detection using the deep learning approach has been the focus of recent studies. This paper presents a preliminary study of high-accuracy fall detection using general machine learning methods, as opposed to intensive and computationally complex deep learning. After pre-processing the radar data, power spectral densities (PSDs) are obtained using the Welch method and used as features in the classifiers. The fall is detected using four different classifiers and their derivatives. The Ensemble bagged trees algorithm detects the falls with 1 0 0% accuracy. The accuracy of other classifier types is also not negligible. In addition, the classification has been carried out with the proposed method for four different types of activities and it has been observed that the results are promising. © 2024 IEEE.eninfo:eu-repo/semantics/closedAccessclassificationfall detectionmachine learningPSDradarContrastive LearningDeep learningPower spectral densityDensity featuresFall detectionHigh-accuracyLearning approachMachine learning methodsMachine-learningPower spectralPre-processingRadar dataWelch methodAdversarial machine learningConference Object10.1109/IDAP64064.2024.107110192-s2.0-85207868179