动态数据下的三支区间离散模型

Three-way Interval Discretization Method Under Dynamic Data

在数据挖掘领域中,数据离散化是将一组连续的数值属性转换为离散的标称属性值,并广泛在音频或视频等连续时间信号的预处理中得到应用.据文献考证,离散区间中的不确定性的空白区间被现有研究所忽略.此外,动态的增量数据将使离散区间更加复杂.针对增量数据下离散问题,本文提出了一种基于三支决策的自适应动态区间离散化方法.本文提出的三支离散化方法可以使离散区间的范围随数据的变化而自适应地变化,并提高了对新增量数据进行离散化的效果.利用本文定义的空白区间的概念,可有效提高新数据和原始数据之间融合的效果.实验结果表明,本文的方法对于处理增量式数据离散化问题具有较好的效果,且运行速度更快.

Discretization is an indispensable processing step when processing continuous-time signals such as audio or video in the field of data mining.The main objective of discretization is to convert a set of continuous numerical attributes into discrete nominal attribute.The blank area among the discrete intervals are ignored by the existing researches as far as we well-know.Besides,the dynamic incremental data will make the discrete intervals more complex.To aim at the incremental continuous data,this paper proposes an adaptive dynamic interval discretization method based on three-way decisions.The novel three-way discretization method proposes an alternative way to make the discrete intervals vary with the variation of data adaptively,and it improves the performance of discretization on the new incremental data.The blank area defined in this paper have good performance in the fusion processing between new and original data.The experimental results show that the proposed method outperforms other well-know discretization and runs faster.