Application of improved attribute recognition model based on non-linear measure function to comprehensive assessment on river water quality

Misjudgements occurred when the general attribute recognition model based on linear measure function(LMF-ARM) is used to make assessment on the virtual samples drawn randomly from the criterion of water quality. This leads to the down-ward reliability of the result while LMF-ARM is applied to the actual water samples. Therefore,the improved attribute recognition model based on non-linear measure function(NLMF-ARM)is proposed here. The results given by the later model are much better than the former,according to the tests of the virtual samples selected by both the random method and the orthogonal design. It indicates that the measure function can play an important role during the process of utilizing attribute recognition model to make comprehensive assessment. So the conclusion can be draw from the case on a city groundwater quality assessment,that non-linear measure function,compared with the linear,has better abilities to describe the natural attribute degree of assessment indexes. Because of the higher reliability than LMF-ARM,NLMF-ARM has wider applicability in the comprehensive assessment of water quality.