Comparison of SVM, Naive Bayes, and ELM Models in Plant Growth Classification

This study investigates the application of machine learning models to predict plant growth milestones based on environmental and treatment data. The dataset comprises categorical variables such as soil type, water frequency, and fertilizer type, alongside numerical variables including sunlight ho...

Full description

Saved in:
Bibliographic Details
Main Authors: M., Muflih, Silvia, Ratna, Haldi, Budiman, Usman, Syapotro, Muhammad, Hamdani
Format: Article
Language:English
English
Published: INTI International University 2024
Subjects:
Online Access:http://eprints.intimal.edu.my/2055/1/jods2024_56.pdf
http://eprints.intimal.edu.my/2055/2/596
http://eprints.intimal.edu.my/2055/
http://ipublishing.intimal.edu.my/jods.html
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study investigates the application of machine learning models to predict plant growth milestones based on environmental and treatment data. The dataset comprises categorical variables such as soil type, water frequency, and fertilizer type, alongside numerical variables including sunlight hours, temperature, and humidity. Preprocessing involved one-hot encoding for categorical variables and standard scaling for numerical features. The models employed were Support Vector Machine (SVM), Naive Bayes, and Extreme Learning Machine (ELM). The baseline SVM model achieved an accuracy of 58.97%, and hyperparameter tuning using GridSearchCV did not improve this performance, maintaining the accuracy at 58.97%. The Naive Bayes model achieved an accuracy of 51.28%, while the ELM model had an accuracy of 43.85%. Among the models, the SVM demonstrated the highest accuracy, though further improvement is required for practical implementation. The findings underscore the importance of selecting appropriate machine learning models and optimizing their parameters to enhance prediction accuracy in agricultural applications. Despite the SVM's superior performance in this context, continued refinement is essential to address the challenges posed by predicting plant growth milestones accurately.