Thai Water Buffalo Disease Analysis with the Application of Feature Selection Technique and Multi-Layer Perceptron Neural Network

• Main Authors: S. Nuanmeesri, W. Sriurai
• Format: Article
• Language: English
• Published: D. G. Pylarinos 2021-04-01
• Series: Engineering, Technology & Applied Science Research
• Subjects: water buffalo diseases; feature selection; multi-layer perceptron; neural network; synthetic minority over-sampling
• Online Access: https://etasr.com/index.php/ETASR/article/view/4049
• ISSN: 2241-4487; 1792-8036

This research aims to develop the analysis model for diseases in water buffalo towards the application of the feature selection technique along with the Multi-Layer Perceptron neural network. The data used for analysis was collected from books and documents related to diseases in water buffalo and the official website of the Department of Livestock Development. The data consists of the characteristics of six diseases in water buffalo, including Anthrax disease, Hemorrhagic Septicemia, Brucellosis, Foot and Mouth disease, Parasitic disease, and Mastitis. Since the amount of the collected data was limited, the Synthetic Minority Over-sampling Technique was also employed to adjust the imbalance dataset. Afterward, the adjusted dataset was used to select the disease characteristics towards the application of two feature selection techniques, including Correlation-based Feature Selection and Information Gain. Subsequently, the selected features were then used for developing the analysis model for diseases in water buffalo towards the use of Multi-Layer Perceptron neural network. The evaluation results of the model’s effectiveness, given by the 10-fold cross-validation, showed that the analysis model for diseases in water buffalo developed by Correlation-based Feature Selection and Multi-Layer Perceptron neural network provided the highest level of effectiveness with the accuracy of 99.71%, the precision of 99.70%, and the recall of 99.72%. This implies that the analysis model is effectively applicable.