基于WEKA的高速公路换卡逃费问题研究

WEKA in Studying Expressway Toll Evasion by Changing Cards

对日益严重的高速公路偷逃通行费现象进行分析,重点研究了常见的换卡逃费问题。选取部分流水数据进行预处理,在MyEclipse平台上使用开源数据挖掘工具WEKA,通过C4.5决策树算法对数据进行分析,建立模型并进行预测。结果表明模型准确率较高,能挖掘隐藏在历史记录中未被发现的换卡逃费记录,还可以对流水数据进行预测并提供置信度,对解决其他类型的偷逃费行为也具有一定的参考价值。

This paper analyses the growing problem of expressway toll evasion, by changing cards particularly. By processing a set of contiguous data on the platform of MyEclipse with C4.5 decision tree algorithm of WEKA, an open source data mining tool, the modeling and prediction of the activities are conducted. The results prove that the model is effective and can mine the historic records to find out events of toll evasion by changing cards. The method may be used for studying toll evasion of other types.