摘要： 提出一种多标记学习阈值确定算法(DTML)，为每个类别标记确定一个阈值。当分类器将一个测试示例预测为某个类别标记的分值大于该类别标记的阈值时，则将该类别标记添加到该测试示例的最终分类结果中。该算法采用编程实现，并将其应用于PT5方法和TML算法。实验结果表明，利用DTML算法为多标记学习算法确定阈值，能够得到较好的分类效果。

关键词: 数据挖掘, 多标记学习, 阈值确定

Abstract: This paper proposes a new algorithm DTML(Determination Threshold for Multi-label Learning). An optimal threshold is set for each label by learning from the training data set. When the score for a test instance to one label is bigger than threshold which is set for the label, the label will be added to the last classifying result for the test instance. This method is implemented and applied to PT5 method and TML algorithm. Experimental results show that the DTML method can get a well performance on learning threshold for multi-label algorithms.

Key words: data mining, multi-label learning, threshold determination

中图分类号: 

• TP301.6