Classification of multivariate time series and structured data using constructive induction
Classification of multivariate time series and structured data using constructive induction
复制标题
DOI：
10.1007/s10994-005-5826-5
复制
发表时间：
2005-02-01
期刊：
影响因子：
7.5
通讯作者：
Sammut, C
中科院分区：
文献类型：
--
作者：
Kadous, MW;Sammut, C
我们提出了一种针对多变量时间序列数据的团队任务的建设性归纳方法。使用元特征，属性值组合的范围扩展到具有某种重复子结构的实例的域，例如手写识别中的笔画，或时间序列数据中的局部最大值。子结构的类型由用户定义，但会自动提取并用于构造属性。元特征应用于两个实际领域：手语识别和心电分类。使用元特征，我们能够生成可理解或准确的分类器，产生可与手工预处理和人类专家相媲美的结果。
We present a method of constructive induction aimed at teaming tasks involving multivariate time series data. Using metafeatures, the scope of attribute-value teaming is expanded to domains with instances that have some kind of recurring substructure, such as strokes in handwriting recognition, or local maxima in time series data. The types of substructures are defined by the user, but are extracted automatically and are used to construct attributes.Metafeatures are applied to two real domains: sign language recognition and ECG classification. Using metafeatures we are able to generate classifiers that are either comprehensible or accurate, producing results that are comparable to hand-crafted preprocessing and comparable to human experts.