Abstract: The comprehensive evaluation method of groundwater quality is of some defects. In this paper, the weight of evaluation indicator difference is obtained by calculating entropy. Using the concept of the ordered partition class, the attribute recognition model based on coefficient of entropy is established. By selecting indicators of TDS, COD, total Fe, total Mn, Cl-, SO42-, NO2- and F-, the shallow groundwater quality of the Southern Suburbs Landfill of Kaifeng is evaluated with the attribute recognition model based on coefficient of entropy. The results show that the water quality evaluated with the attribute recognition model based on coefficient of entropy is better than the results of the comprehensive evaluation; the weights of SO42-and NO2- are maximum, and the weights of Cl-, COD, total Fe and F- are followed, while the weights of TDS and total Mn are minimum; the proportion of the distribution of theⅠ, Ⅱ, Ⅲ, Ⅳ and Ⅴclass-type water in the study area is 1.7%, 2.5% , 26.7%, 12.2% and 56.9%, respectively; the distribution of shallow groundwater quality in the study area has a certain regularity, the water quality of Well L01 and Well G54 close to the landfill is the worst Ⅴclass-type water. As this center extended outward, the water quality gradually improved, and the Ⅲ and Ⅳ class-type water distributed in turn. The distribution of Ⅲ class-type water migrated to the north and east of the study area, while theⅠand Ⅱ class-type water almost distributed in the south of the study area.