数据挖掘技术在水质评价中应用研究  

Study on Water Quality Evaluation Based on Data mining

作　　者：邹国平[1] 彭梅香[1] 

机构地区：[1]新余学院计算机系,江西新余338000

出　　处：《计算机仿真》2012年第1期148-151,共4页Computer Simulation

基　　金：国家自然科学基金项目(70971131)

摘　　要：研究水质评价准确性问题,提高水质评价结果。水质评价指标多,根据监测资料,每种指标对评价结果产生不同程度影响,即权重值不同。传统评估采用人为确定指标权重,具有盲目性和主观性,导致水质评价结果不科学,准确率低。为了提高水质评价的准确率,提出一种基于熵权的水质评价模型。首先构建水质评价指标体系,然后利用熵权法计算出各评价指标权重,最后利用逼近理想解排序法计算出各检测点到理想解和负理想解的距离,参照水质污染标准,得到各检测点水质污染等级,利用相对贴近度,获得水质污染值。研究结果表明,评价准确度高,是一种有效的水质评价方法,为评价水质提供了依据。Research about water quality evaluation. The traditional methods of water quality evaluation are influenced by experts＇ subjective experiences, the evaluation accuracy is not high, and the evaluation result explicable is very poor. In order to improve the water quality evaluation accuracy, the paper put forward a water quality evaluation model based on approximate entropy weight and TOPSIS. Firstly, the water quality evaluation index system was constructed. Secondly, the entropy weight method was used to calculate the weight of each contestant indexes. Finally, TOPSIS was used to calculate the various testing point, combined with ideal solution positive and minus ideal solutions distance. Refering to the water pollution standard, each testing point was obtained. Combined with water pollution levels, relative closeness degrees were used to obtain water pollution values. Research results show that the model of the evaluation result is more precise than fuzzy evaluation, the evaluation illustrate is perfect, and the model is an effective water quality evaluation model.

关 键 词：熵权法 逼近理想解排序方法 水质评价 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]