Abstract: In order to overcome the shortage of the traditional land evaluation methods and improve the intelligibility and efficiency of the land evaluation knowledge through which evaluation results could better reflect the changes in land with the essential characteristics and the natural law, a land evaluation method combining association rule with fuzzy decision was proposed in this paper. Database was selected from the database of Second General Soil Survey Database of Guangdong Province. The experimental results demonstrate that the proposed method is convenient to extract intelligible land evaluation association rules. When setting the minimum support at 0.005, 0.003 and 0.001, 32, 54 and 126 land evaluation rules can be extracted, respectively, and by using only 126 rules, 95.25% of area correct rate of land evaluation can be obtained.