Open AccessData Mining Approach to Herbs Classification
Author(s) -
Adillah Dayana Ahmad Dali,
Nurul Aswa Omar,
Aida Mustapha
Publication year - 2018
Publication title -
indonesian journal of electrical engineering and computer science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.241
H-Index - 17
eISSN - 2502-4760
pISSN - 2502-4752
DOI - 10.11591/ijeecs.v12.i2.pp570-576
Subject(s) - decision tree , variety (cybernetics) , identification (biology) , computer science , artificial intelligence , data mining , machine learning , artificial neural network , table (database) , tree (set theory) , mathematics , botany , biology , mathematical analysis
Herbs are one of the high-value products in Malaysia. The term ‘herbs’ has more than one definition. It is also demanding by multiple manifolds. Herbs are used in many sectors nowadays. The ability to identify variety herbs in the market is quite hard without the intervention of human experts. Unfortunately, human experts are prone to error. Herbs classification is able to assist human experts and at the same time minimizing the intervention. This research performs identification and classification of herbs based on image capture ad variety of classification algorithms such as an Artificial Neural Network (ANN), K-Nearest Neighbors (IBK), Decision Table (DT) and M5P Tree algorithms. The selected algorithms are implemented and evaluated to their relative performance and IBK is found to produce the highest quality outputs.