Downloads: 1 | Views: 319 | Weekly Hits: ⮙1 | Monthly Hits: ⮙1
Research Paper | Computer Engineering | India | Volume 12 Issue 9, September 2023 | Popularity: 5.2 / 10
Design of Improved Metaheuristics with Machine Learning Driven Human Activity Recognition
L. Maria Anthony Kumar, Dr. S. Murugan, A. Therasa Alphonsa
Abstract: Human Activity Recognition (HAR) is a significant area of research and application within the fields of computer vision, machine learning, and wearable technology. It involves the development of algorithms and models to automatically identify and classify human activities dependent upon data gathered in several sensors like accelerometers, gyroscopes, and even video cameras. The goal of HAR is to enable computers and systems to understand and interpret human actions and movements.In recent years, DL approaches have shown remarkable performance in HAR tasks. With this motivation, this study designs an improved metaheuristics with machine learning driven human activity recognition approach (IMML-HARA). The IMML-HARA technique focuses on the recognition and classification of human activities. In the presented IMML-HARA technique, Improved Chicken Swarm Optimization (ICSO) Algorithm for electing an optimal set of features. Moreover, the IMML-HARA technique offers the design of radial basis function (RBF) classifier for the identification of human activities into distinct activities. The performance of the IMML-HARA technique is tested on two benchmark HAR datasets. The experimental results indicate the betterment of the IMML-HARA method over other existing recent state of art approaches.
Keywords: Machine Learning; Radial Basis Function; Chicken Swarm Optimization; Human Activity Recognition
Edition: Volume 12 Issue 9, September 2023
Pages: 141 - 148
DOI: https://www.doi.org/10.21275/SR23822120407
Please Disable the Pop-Up Blocker of Web Browser
Verification Code will appear in 2 Seconds ... Wait