A Novel Hybrid Gradient-Based Optimizer and Grey Wolf Optimizer Feature Selection Method for Human Activity...

Area of Science:

• Computer Science
• Artificial Intelligence
• Machine Learning

Background:

• Human Activity Recognition (HAR) is crucial for applications like elderly care, smart homes, and healthcare monitoring.
• High-dimensional data in HAR systems often degrades model performance.
• Existing methods struggle with the complexity of HAR data.

Purpose of the Study:

• To propose an efficient HAR system using a lightweight feature selection method.
• To enhance HAR classification accuracy by addressing data dimensionality.
• To introduce the GBOGWO feature selection algorithm.

Main Methods:

• Developed a hybrid feature selection (FS) method named GBOGWO, combining Gradient-based Optimizer (GBO) and Grey Wolf Optimizer (GWO).
• Employed GBOGWO for optimal feature selection in HAR datasets.
• Utilized Support Vector Machine (SVM) for activity classification post-feature selection.

Main Results:

• The GBOGWO method demonstrated superior performance in feature selection for HAR.
• Achieved an average classification accuracy of 98% on the UCI-HAR and WISDM datasets.
• The proposed system effectively handles high-dimensional data for improved HAR.

Conclusions:

• GBOGWO is an effective feature selection technique for enhancing HAR systems.
• The integration of GBO and GWO operators provides a robust approach to feature selection.
• The developed HAR system offers high accuracy and efficiency for real-world applications.