Pallet-level Classification Using Principal Component Analysis in Ensemble Learning Model

Authors

  • Chun Sern Choong
  • Ahmad Fakhri Ab. Nasir
  • Muhammad Aizzat Zakaria
  • Anwar P.P. Abdul Majeed
  • Mohd Azraai Mohd Razman Universiti Malaysia Pahang

DOI:

https://doi.org/10.15282/mekatronika.v2i1.6720

Keywords:

Pallet-level, RFID, Ensemble Learning, Features Selection, RSSI

Abstract

In this paper, we present a machine learning pipeline to solve a multiclass classification of radio frequency identification (RFID) signal strength. The goal is to identify ten pallet levels using nine statistical features derived from RFID signals and four various ensemble learning classification models. The efficacy of the models was evaluated by considering features that were dimensionally reduced via Principal Component Analysis (PCA) and original features. It was shown that the PCA reduced features could provide a better classification accuracy of the pallet levels in comparison to the selection of all features via Extra Tree and Random Forest models.

Downloads

Published

2020-06-05

How to Cite

[1]
C. S. Choong, A. F. Ab. Nasir, M. A. Zakaria, A. P.P. Abdul Majeed, and M. A. Mohd Razman, “Pallet-level Classification Using Principal Component Analysis in Ensemble Learning Model”, Mekatronika: J. Intell. Manuf. Mechatron., vol. 2, no. 1, pp. 23–27, Jun. 2020.

Issue

Section

Original Article

Most read articles by the same author(s)

<< < 1 2 3 4