Human Activity Recognition Using a Hybrid Approach of Radial Basis Neural Networks and Support Vector Machines

The Internet of Things (IoT) has become increasingly prevalent, and recent advances in machine learning, particularly in healthcare, have gained significant attention from researchers. One prominent interdisciplinary topic in these fields is human activity recognition (HAR). Despite extensive research, several challenges remain in this area, especially concerning the application of modern machine learning techniques for HAR. This study proposes a novel method for human activity recognition by combining radial basis function neural networks (RBFNN) and support vector machines (SVM). The approach enhances recognition accuracy and algorithm efficiency by extracting relevant features using RBFNN and convolutional neural networks (CNN). Classification is then performed using SVM. The proposed method was evaluated using the UCI HAR dataset, which includes six distinct human activities. Results demonstrate that the proposed approach achieves an accuracy of 99%, surpassing existing methods.