Comparing Performances of Data Mining Algorithms for Classification of Green Coffee Beans
Edwin R. Arboleda
Edwin R. Arboleda, Department of Computer and Electronics Engineering, Cavite State University (CvSU), Indang, Cavite, Philippines.
Manuscript received on 18 June 2019 | Revised Manuscript received on 25 June 2019 | Manuscript published on 30 June 2019 | PP: 1563-1567 | Volume-8 Issue-5, June 2019 | Retrieval Number: E7331068519/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: An image processing technique extracted four features of green coffee beans of different species which are the liberica, robusta, and excelsa. The acquired datasets were subjected to the Classification Learner App of MATLAB 2017. Among the 22 classifiers included in the Classification Learner App, 94.1 percent is the highest accuracy, obtained by the Coarse Tree Algorithm, while the lowest classification percentage was obtained by Boosted Tree Algorithm with 28.2% accuracy. Out of the 22 algorithms, only 4 got a classification accuracy that is lower than 90 percent and 18 algorithms got an accuracy of more than 90 percent. It can be concluded that the combined image processing and classification using data mining algorithms can be used in classifying green coffee bean species.
Keywords: Artificial Neural Network, Classification Learner App, Image Processing, Morphological Features.
Scope of the Article: Classification