Experimental Evaluation of Time-Series Decision Tree
Experimental Evaluation of Time-Series Decision Tree
复制标题
DOI：
10.1007/11423270_11
复制
发表时间：
2003-10
期刊：
影响因子：
--
通讯作者：
Yuu Yamada;Einoshin Suzuki;H. Yokoi;K. Takabayashi
中科院分区：
文献类型：
--
作者：
Yuu Yamada;Einoshin Suzuki;H. Yokoi;K. Takabayashi
在本文中，我们对各种条件下的时间序列决策树归纳方法进行了实验评估。我们的时间序列树在其内部节点中有一个时间序列属性的值（即时间序列），并根据一对时间序列之间的不相似性来分割示例。我们的方法通过基于类和形状信息的穷举搜索来选择数据中存在的时间序列进行分割测试。根据经验观察，该方法在时间序列分类中引入了准确且全面的决策树，由于其在各种实际应用中的重要性而受到越来越多的关注。该评估揭示了几个重要的发现，包括对比测试与其优度衡量标准之间的相互作用。
In this paper, we give experimental evaluation of our time-series decision tree induction method under various conditions. Our time-series tree has a value (i.e. a time sequence) of a time-series attribute in its internal node, and splits examples based on dissimilarity between a pair of time sequences. Our method selects, for a split test, a time sequence which exists in data by exhaustive search based on class and shape information. It has been empirically observed that the method induces accurate and comprehensive decision trees in time-series classification, which has gaining increasing attention due to its importance in various real-world applications. The evaluation has revealed several important findings including interaction between a split test and its measure of goodness.