Comparative of SVM and Decision Tree Techniques for Predicting Hydroponic Tomato Growth and Yield Using Deep Water Culture

The purpose of this research is to evaluate the effectiveness of SVM and Decision Tree techniques in the context of hydroponic tomato cultivation using the Deep Water Culture (DWC) method. We assess the prediction power of various machine learning techniques with respect to tomato growth, yield, and quality characteristics using a variety of inputs relevant to hydroponic systems. We use the DWC approach to collect data from hydroponic tomato farms. The collection includes a wide range of environmental parameters, including pH values, water temperature, illumination, and nutrient concentrations. It also contains measures unique to plants, like yield, fruit size, and leaf area. The dataset's pertinent features, which include growth characteristics, environmental factors, and nutrient levels, are found. After processing, these chosen features are prepared for model training. SVM and Decision Tree techniques are tested in a hydroponic Deep Water Culture (DWC) system to predict tomato growth, production, and quality. The pH, temperature, illumination, and nutrient levels of hydroponic tomato farms were recorded, along with plant parameters like yield, fruit size, and leaf area. Feature selection and preprocessing prepared data for model training. Decision Tree and SVM predicted tomato production metrics well. SVM had better precision in some circumstances, while Decision Tree was easier to interpret and apply. These algorithms can optimize hydroponic tomato growing, and future research should use ensemble methods to improve forecast accuracy and reliability.