<![CDATA[Agricultural land suitability assessment plays an important role in supporting agricultural planning and ensuring that land resources are used according to environmental characteristics and crop requirements. In Curah Kalak Village, Situbondo Regency, land suitability assessment is generally conducted manually, resulting in limited efficiency and difficulties in accessing spatial information for decision-making. This study aims to develop a WebGIS-based agricultural land suitability classification system by integrating the K-Nearest Neighbor (KNN) algorithm with Geographic Information System (GIS) technology. The classification process uses environmental and spatial parameters, including slope, soil pH, soil type, rainfall, altitude, soil depth, and irrigation distance. A total of 669 agricultural land records were used as the dataset, and land suitability classes were classified using KNN with K = 31 based on Euclidean distance. The developed system classified land suitability for sugarcane, chili, corn, and rice commodities. The classification results indicated that chili and sugarcane were categorized as Highly Suitable (S1), whereas corn and rice were categorized as Moderately Suitable (S2). Performance evaluation using an 80:20 train-test split showed that the KNN model achieved an accuracy of 73.00%, weighted precision of 68.00%, weighted recall of 73.00%, and weighted F1-score of 70.00%, indicating moderate classification performance. Furthermore, the WebGIS provides interactive digital maps for visualizing classification results and spatial information. Black-box testing confirmed that all implemented system functions operated according to the specified requirements. The novelty of this study lies in the integration of KNN-based multi-commodity land suitability classification with WebGIS visualization within a village-level platform for agricultural land assessment. The proposed system can support agricultural land management and spatially informed decision-making in Curah Kalak Village.]]>
