<![CDATA[Changes in various environmental parameters during the cultivation process can have an impact on plant development and have the potential to reduce the quality of the results obtained. Therefore, a method is needed that is able to identify plant conditions quickly, objectively, and accurately. This study aims to build a classification model of chili plant conditions using the Decision Tree J48 algorithm by utilizing environmental parameter data and soil nutrients as the basis for determining healthy and unhealthy plant categories. This research method uses a quantitative approach with data mining techniques on a public dataset consisting of 2,200 datasets. The variables analyzed included nitrogen (N), phosphorus (P), potassium (K), temperature, humidity, soil pH (ph), and rainfall. The process of forming and testing the model was carried out using the WEKA application with the 10-fold cross validation technique. The results showed that moisture was the most dominant attribute in the formation of decision trees, followed by rainfall, temperature, and soil pH as supporting attributes. The resulting model obtained an accuracy rate of 99.5909%, a Kappa Statistic value of 0.989, a Mean Absolute Error (MAE) of 0.0042, and a Root Mean Squared Error (RMSE) of 0.064. These values indicate that the model has an excellent degree of conformity with a relatively low rate of prediction error. Based on these results, the Decision Tree J48 algorithm has proven to be effective in classifying the condition of chili plants based on environmental parameters and has the potential to be applied as the basis for the development of a decision support system to help the process of monitoring plant conditions more quickly, objectively, and accurately.]]>
Copyrights © 2026
