<![CDATA[Soil damage is an important issue in the agricultural sector because it directly affects crop productivity and success. Inaccurate identification of soil damage can lead to errors in crop selection, resulting in suboptimal agricultural yields. This study aims to implement the backward chaining method as an inference engine in an expert system to diagnose soil damage and determine optimal crop planting recommendations. The backward chaining method is used with a goal-based reasoning approach, where the system starts the inference process from the hypothesis of soil damage type and traces the facts in the form of symptoms provided by the user. The system's knowledge base is organized in the form of rules obtained from literaturestudies and expert knowledge in the field of agriculture. The results show that the expert system built is capable of identifying the type of soil damage based on the symptoms entered and providing crop recommendations appropriate to the soil conditions. This system is expected to assist farmers and related parties in making more accurate, effective, and knowledge-based planting decisions.]]>
Copyrights © 2026
