<![CDATA[Drinking water quality is an important factor affecting public health. Accurate water potability classification is essential to ensure safe water consumption. This study aims to compare the performance of Artificial Neural Network (ANN) and Random Forest algorithms for drinking water potability classification using the Water Potability dataset. The dataset consists of 3,276 samples with nine water quality parameters, including pH, hardness, solids, chloramines, sulfate, conductivity, organic carbon, trihalomethanes, and turbidity. Data preprocessing involved missing value handling, normalization using MinMaxScaler, and train-test splitting with 70:30 and 80:20 scenarios. Model performance was evaluated using accuracy, precision, recall, and F1-score metrics. Experimental results show that the 80:20 train-test split produced the best performance for both models. Under this scenario, ANN achieved an accuracy of 0.66, precision of 0.67, recall of 0.88, and F1-score of 0.76, while Random Forest achieved an accuracy of 0.66, precision of 0.66, recall of 0.87, and F1-score of 0.75. The results showed that ANN achieved slightly higher precision, recall, and F1-score values than Random Forest. However, the performance difference between the two models was relatively small, so both models can be considered competitive in classifying drinking water quality.]]>
Copyrights © 2026
