Along with the popularity of computer and rapid development of information technology, how to increase the
accuracy of the agricultural diagnosis becomes a difficult problem of popularizing the agricultural expert system.
Analyzing existing research, baseing on the knowledge acquisition technology of rough set theory, towards great sample
data, we put forward a intelligent diagnosis model. Extract rough set decision table from the samples property, use
decision table to categorize the inference relation, acquire property rules related to inference diagnosis, through the
means of rough set knowledge reasoning algorithm to realize intelligent diagnosis. Finally, we validate this diagnosis
model by experiments. Introduce the rough set theory to provide the agricultural expert system of great sample data a
effective diagnosis model.