DTW-D: time series semi-supervised learning from a single example
DTW-D: time series semi-supervised learning from a single example
复制标题
DOI：
10.1145/2487575.2487633
复制
发表时间：
2013-08
期刊：
影响因子：
--
通讯作者：
Yanping Chen;Bing Hu;Eamonn J. Keogh;Gustavo E. A. P. A. Batista-Gustavo-E.-A.-P.-A.-Batista-145666101
中科院分区：
文献类型：
--
作者：
Yanping Chen;Bing Hu;Eamonn J. Keogh;Gustavo E. A. P. A. Batista-Gustavo-E.-A.-P.-A.-Batista-145666101
时间序列数据的分类是一个重要的问题，几乎在每一个科学奋进的应用。致力于时间序列分类的大型研究社区通常使用UCR Archive来测试他们的算法。在这项工作中，我们认为，这种资源的可用性隔离了大部分的研究社区从以下现实，标记的时间序列数据往往是很难获得。这个问题的明显解决方案是半监督学习的应用;然而，正如我们将要展示的那样，直接应用现成的半监督学习算法通常不适合时间序列。在这项工作中，我们解释了为什么半监督学习算法通常在时间序列问题上失败，我们介绍了一个简单但非常有效的解决方案。我们展示了我们的想法在不同的真实的字的问题。
Classification of time series data is an important problem with applications in virtually every scientific endeavor. The large research community working on time series classification has typically used the UCR Archive to test their algorithms. In this work we argue that the availability of this resource has isolated much of the research community from the following reality, labeled time series data is often very difficult to obtain. The obvious solution to this problem is the application of semi-supervised learning; however, as we shall show, direct applications of off-the-shelf semi-supervised learning algorithms do not typically work well for time series. In this work we explain why semi-supervised learning algorithms typically fail for time series problems, and we introduce a simple but very effective fix. We demonstrate our ideas on diverse real word problems.