Article: An enhanced multi-kernel-based extreme learning machine model for crop yield prediction in IoT-based smart agriculture Journal: International Journal of System of Systems Engineering (IJSSE) 2024 Vol.14 No.5 pp.504 - 519 Abstract: Smart agriculture is a terrific approach to boosting agricultural output and farm productivity, whereas the Internet of Things (IoT) provides production and control facilities with intelligent navigation. Large-scale physical surveys and the use of remote sensing data are two approaches that are widely used for crop prediction. Due to the growing volume of data generated by remote sensing images and the requirement for more sophisticated algorithms to identify the underlying spatiotemporal patterns of this data, this approach is essential for the issue of forecasting agricultural yields. Despite the fact that this field has made great strides owing to machine learning techniques, here, we suggest a machine learning-based automated prediction approach. The crop production can be accurately predicted by the suggested optimised multi-kernel-based extreme learning machine (ELM) model. By employing the adaptive rat optimisation technique to optimise the kernel parameters of kernel functions, the performance of the multi-kernel-based ELM is improved in this detection model. The recommended OMK-ELM model can detect crop yield output in IoT agriculture with a maximum accuracy of 98.462%, precision of 93.627%, and recall of 99.721%, according to testing results. Inderscience Publishers - linking academia, business and industry through research

Title: An enhanced multi-kernel-based extreme learning machine model for crop yield prediction in IoT-based smart agriculture

Authors: Yogomaya Mohapatra; Anil Kumar Mishra

Addresses: Department of Computer Science and Engineering, C.V. Raman Global University, Bhubaneswar, 752054, Odisha, India ' Department of AIML, Gandhi Engineering College, Bhubaneswar, 752054, Odisha, India

Abstract: Smart agriculture is a terrific approach to boosting agricultural output and farm productivity, whereas the Internet of Things (IoT) provides production and control facilities with intelligent navigation. Large-scale physical surveys and the use of remote sensing data are two approaches that are widely used for crop prediction. Due to the growing volume of data generated by remote sensing images and the requirement for more sophisticated algorithms to identify the underlying spatiotemporal patterns of this data, this approach is essential for the issue of forecasting agricultural yields. Despite the fact that this field has made great strides owing to machine learning techniques, here, we suggest a machine learning-based automated prediction approach. The crop production can be accurately predicted by the suggested optimised multi-kernel-based extreme learning machine (ELM) model. By employing the adaptive rat optimisation technique to optimise the kernel parameters of kernel functions, the performance of the multi-kernel-based ELM is improved in this detection model. The recommended OMK-ELM model can detect crop yield output in IoT agriculture with a maximum accuracy of 98.462%, precision of 93.627%, and recall of 99.721%, according to testing results.

Keywords: crop yield prediction; ELM; extreme learning machine; IoT; Internet of Things; OMK-ELM; adaptive rat optimisation algorithm; machine learning.

DOI: 10.1504/IJSSE.2024.140755

International Journal of System of Systems Engineering, 2024 Vol.14 No.5, pp.504 - 519

Received: 19 Jan 2023
Accepted: 30 Jan 2023
Published online: 02 Sep 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: An enhanced multi-kernel-based extreme learning machine model for crop yield prediction in IoT-based smart agriculture

Keep up-to-date