Classification of Sugar-Cane Varieties' Disease Resistance using Bagging and Boosting Machine Learning Algorithms
Abstract
Sugarcane, a vital agricultural produce, is susceptible to diseases that can severely impact crop quality and yield. Early identification and effective mitigation of sugarcane diseases are crucial for successful crop management. Disease outbreaks can lead to significant financial losses for farmers as they can devastate entire crop fields. To combat this issue, researchers are exploring the application of Artificial Intelligence (AI) techniques, particularly Bagging and Boosting algorithms in Machine Learning (ML), to analyze agricultural data and prevent crop damage caused by various factors, with diseases being a major concern.
The research is prompted by the rapid evolution of sugarcane disease classes and the lack of disease diagnostic and recognition skills among farmers. By leveraging Bagging and Boosting algorithms, sugarcane farmers can gain valuable insights into disease identification and prediction, enabling them to take timely preventive measures. Integrating AI and ML techniques in sugarcane cultivation can enhance disease management strategies, safeguard crop productivity, and contribute to sustainable agricultural practices.
Keywords
Download Options
Introduction
Sugarcane sicknesses are a colossal wellspring of concern and risk for ranchers, since they can monetarily affect sugarcane result and creation, in the event that they are not recognized as soon a possible [5]. In the event that the development of a particular yield declines, it adversely affects the economy. Supportability underway is fundamental, as is asset productivity for seeds, water, soil, and manures. On the off chance that these harvests are annihilated while creating, horticultural produce alongside the expectation of keeping up with nature of these yields will lose a portion of their seriousness. Since the sugarcane infections are inescapable, it is basic to perceive and analyze them. Sugarcane plant sicknesses are an enormous logical subject of study that spotlights on the illness' natural elements. Plant illness location and determination has shown to be intriguing and needs specific thought [7]. Sanitation is influenced by plant infections and the illnesses are especially destructive to limited scope ranchers who depend on sufficient result to get by. Identifying these contaminations at beginning phases will bring about better sugarcane creation, and will help the two ranchers and clients. Early location of sugarcane diseases is used to execute protection estimates to limit extra damage.
Conclusion
In this study, we explored the application of two powerful machine learning algorithms, Bagging and Ada Boosting, for predicting sugarcane disease resistance levels. The experimental results demonstrated the effectiveness of both algorithms in accurately classifying sugarcane varieties into resistant, intermediate, and susceptible categories. Additionally, the Ada Boosting algorithm exhibited superior performance compared to Bagging, indicating its potential as a robust predictive tool for sugarcane disease resistance classification.
The high accuracy, precision, and recall scores obtained from both models highlight their ability to make accurate predictions and minimize false positives and false negatives. Such performance is critical in the context of sugarcane disease management, as misclassifications could lead to improper disease control measures and impact crop productivity.
It is worth noting that while both Bagging and Ada Boosting algorithms performed well in this study, the choice of algorithm depends on specific requirements, such as computational resources and model interpretability. Further investigations on larger and diverse datasets are warranted to confirm the generalizability of these results and to explore other potential machine learning approaches.