Analyzing Hydroponic Farming with a Drip System and Comparing it to a Decision Tree (DT) using a Logistic Regression (LR)

This study will assess Decision Tree and Support Vector Machine (SVM) algorithms’ ability to predict tomato quality, yield, and growth in Deep Water Culture (DWC) hydroponic farming. We evaluate multiple machine-learning methods for hydroponic input prediction. The DWC approach collects data from hydroponic tomato farms. Yield, fruit size, leaf area, $\mathbf{p H}$, water temperature, illumination, and fertilizer concentrations are among these statistics. Key dataset features like growth attributes, environmental impacts, and nutrient concentrations are selected for model training. The result reveals that Decision Tree and SVM can forecast tomato production and growth factors. SVM may be more accurate than a Decision Tree, although the latter is simpler to understand. In tomato yield prediction, SVM is $\mathbf{8 8 \%}$ accurate and Decision Tree $\mathbf{8 5 \%}$. The predicted parameters affect recall and precision; SVM usually has higher precision. Conclusion: Decision Trees and Support Vector Machines can forecast tomato farming results in DWC hydroponic systems. Decision Tree is simpler and interpretable, although SVM is more precise in some cases.