基于聚类的交易者煤炭质量检验可信度研究

Cluster-based Research on Believable Level of Coal Quality Test of Traders

研究提出了一个煤炭质量检验可信度的新定义,并设计了应用可信度识别交易者质量检验行为的聚类计算模型。在运用某大型煤炭企业及其联盟用户的煤炭质量检验数据实证基础上,研究发现:交易者煤炭质量检验行为存在统计规律,聚类方法能够在统计意义下识别交易者的质量检验行为特征,并据此推断商品的质量可信度。

We present a new definition for coal quality test believable level in this paper, which is useful to develop a clustering model to identify traders＇ quality test behaviors. Based on the empirical analysis about the coal quality test data from a large-scale firm and its alliance users, we find that coal traders quality test behavior exist some statistical regularity, and that under statistical significance the clustering method can be used to classify the traders according to their behavioral characteristics from which the quality believable level can be deduced.