Cocoa production in Indonesia encounters ongoing challenges due to declining plantation areas and suboptimal land utilization. This study applies the ID3 Spatial algorithm to evaluate land suitability for cocoa cultivation in Bogor Regency, West Java Province. The methodology integrates nine basic land characteristics, including elevation, drainage, relief, base saturation, cation exchange capacity, soil texture, soil pH, and mineral soil depth, derived from field surveys conducted by BBSDLP. Two classification models were developed and tested using spatial data preprocessing techniques. Model M1 was the baseline approach without constraints, while Model M2 incorporated a minimum planted area threshold of ≥1 ha. The results show notable performance differences between models. Model M2 achieved a reasonable accuracy of 87.27% compared to Model M1’s 29.09%, with relief identified as the root node due to its higher gain value and reduced entropy. Classification results indicate that Bogor Regency’s cocoa cultivation potential comprises 16,443 ha of S2 (moderately suitable) land and 231,018 ha of S3 (marginally suitable) land. The generated land suitability map may provide stakeholders with helpful guidance for identifying potential cultivation areas. The result suggests that artificial intelligence integration, specifically the ID3 spatial algorithm, could improve land suitability evaluation processes, potentially supporting more informed agricultural development decisions.
Nurkholis et al. (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: