Human Activity Recognition for Elderly People Using Machine and Deep Learning Approaches

Abstract

There are more than 962 million people aged 60 and up globally. Physical activity declines as people get older, as does their capacity to undertake everyday tasks, effecting both physical and mental health. Many researchers use machine learning and deep learning methods to recognize human activities, but very few studies have been focused on human activity recognition of elderly people. This paper focuses on providing assistance to elderly people by monitoring their activities in different indoor and outdoor environments using gyroscope and accelerometer data collected from a smart phone. Smart phones have been routinely used to monitor the activities of persons with impairments; routine activities such as sitting, walking, going upstairs, going downstairs, standing, and lying are included in the dataset. Conventional Machine Learning and Deep Learning algorithms such as k-Nearest Neighbors, Random Forest, Support Vector Machine, Artificial Neural Network, and Long Short-Term Memory Network are used for human activity recognition. Long Short-Term Memory is a recurrent neural network variation that is best suited to handling temporal sequences. Two-fold and ten-fold cross-validation methods were performed to show the effect of changing the data in the training and testing dataset. Among all the classification techniques, the proposed Long Short-Term Memory Network gave the best accuracy of 95.04%. However, Support Vector Machine gave 89.07% accuracy with a very low computational time of 0.42 min using 10-fold cross-validation.

A Systematic Literature Review on Human Activity Recognition

“…In the previous years, sensor technology has made remarkable strides in several areas including computational power and manufacturing costs with accuracy and size [19]. These developments have facilitated the integration of various sensors into smartphones and portable devices, enhancing their intelligence and utility.…”

“…HAR finds applications in behavior analysis [10], surveillance systems [8,9], gesture recognition [11][12][13], Ambient Assisted Living (AAL) [16,17], various healthcare systems [18,19] and patient monitoring [14,15]. For instance, patients with chronic conditions need to adhere to specific diets and exercise regimens [20].…”

“…This model shares parameters between user recognition and activity modules, improving activity recognition through a common attention device. Sheng et al [19] introduced inadequately supervised multi-task depiction, utilizing Siamese networks as well as a temporal backbone model using the convolutional network. According to Bai et al [20], a discriminative adversarial multiview network employing CNN extract multi-view feature utilizing WGAN and Siamese network design to reduce deviations between various subjects' features.…”

Human Activity Recognition for Elderly People Using Pseudonym-Based Conditional Privacy-Preservation Authentication

“…Healthcare systems face major challenges from this growth, as aging is associated with decreased physical activity, impacting both physical and mental health. As more and more elderly people suffer from age-related diseases and malfunctions of body parts, the demand for intelligent health support systems is increasing year by year [1][2][3]. Because of this, there is a need to create a monitoring activity recognition that is used as an innovation that supports the well-being of the elderly [4].…”

Piece-Wise Linear Chaotic Mapping-based Beluga Whale Optimization Algorithm-based Indoor Activity Monitoring for Elderly and Visually Impaired Persons

“…This hybrid method is run on top of a fog-based cloud structure, fogbus. In Hayat et al (2022), the focus was on supporting elderly persons by observing their actions in diverse outdoor and indoor surroundings by means of accelerometer data and gyroscope data gathered from smart phones. LSTM refers to a RNN variation that can be suitable to manage temporal series.…”