Proceedings of the 11th International Applied Business and Engineering Conference, ABEC 2023, September 21st, 2023, Bengkalis, Riau, Indonesia

Research Article

Machine Learning Approaches for Fish Pond Water Quality Classification: Random Forest, Gaussian Naive Bayes, and Decision Tree Comparison

Download32 downloads
  • @INPROCEEDINGS{10.4108/eai.21-9-2023.2342964,
        author={Danuri  Danuri and Muhammad Syafiq Mohd Pozi},
        title={Machine Learning Approaches for Fish Pond Water Quality Classification: Random Forest, Gaussian Naive Bayes, and Decision Tree Comparison},
        proceedings={Proceedings of the 11th International Applied Business and Engineering Conference, ABEC 2023, September 21st, 2023, Bengkalis, Riau, Indonesia},
        publisher={EAI},
        proceedings_a={ABEC},
        year={2024},
        month={2},
        keywords={fish pond water quality classification machine learning},
        doi={10.4108/eai.21-9-2023.2342964}
    }
    
  • Danuri Danuri
    Muhammad Syafiq Mohd Pozi
    Year: 2024
    Machine Learning Approaches for Fish Pond Water Quality Classification: Random Forest, Gaussian Naive Bayes, and Decision Tree Comparison
    ABEC
    EAI
    DOI: 10.4108/eai.21-9-2023.2342964
Danuri Danuri1,*, Muhammad Syafiq Mohd Pozi2
  • 1: Politeknik Negeri Bengkalis
  • 2: Universiti Utara Malaysia
*Contact email: danuri@polbeng.ac.id

Abstract

The health and production of fish in fish farms are greatly influenced by the water quality. This study examines three Machine Learning(ML) methods for categorizing fish pond water quality: Random Forest(RF), Gaussian Naive Bayes(GNB), and Decision Tree(DT). Accuracy, precision, recall, and the F1-Score as a performance indicator are taken into account while evaluating the model. The evaluation findings reveal that RF and GNB outperform DT in every evaluation criteria. GNB, with a rating of 0.958932, had the highest accuracy, followed by RF, with a value of 0.955822, and DT, with a value of 0.932269. The consistent performance of GNB and RF in precision, recall, and F1-Score underscores their superiority.