2014
DOI: 10.1007/s10618-014-0380-z
|View full text |Cite
|
Sign up to set email alerts
|

Classification-driven temporal discretization of multivariate time series

Abstract: Biomedical data, in particular electronic medical records data, include a large number of variables sampled in irregular fashion, often including both time point and time intervals, thus providing several challenges for analysis and data mining. Classification of multivariate time series data is a challenging task, but is often necessary for medical care or research. Increasingly, temporal abstraction, in which a series of raw-data time points is abstracted into a set of symbolic time intervals, is being used … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
55
0

Year Published

2015
2015
2021
2021

Publication Types

Select...
8
1

Relationship

3
6

Authors

Journals

citations
Cited by 92 publications
(55 citation statements)
references
References 37 publications
0
55
0
Order By: Relevance
“…These models are listed in Table 3, along with their references. Note that, though Random Forest (RF) was originally designed for classification problems, here, the range of the class variable is discretized so that such an algorithm can be adopted for regression problems [93]. …”
Section: Methodsmentioning
confidence: 99%
“…These models are listed in Table 3, along with their references. Note that, though Random Forest (RF) was originally designed for classification problems, here, the range of the class variable is discretized so that such an algorithm can be adopted for regression problems [93]. …”
Section: Methodsmentioning
confidence: 99%
“…There are also algorithms that transform time series into simpler representations while aiming at preserving some of the underlying temporal order. These algorithms include Discrete Fourier Transformation (DFT) [17], Discrete Wavelet Transformation (DWT) [18], Symbolic Aggregate approXimation (SAX) [19,20], Equal Width Discretization (EWD), Equal Frequency Discretization (EFD), Persist [21] and Temporal Discretization for Classication (TD4C) [22]. The SAX representation is often preferred since it is intuitive and easy to comprehend while maintaining competitive performance against other representations.…”
Section: Introductionmentioning
confidence: 99%
“…The SAX representation is often preferred since it is intuitive and easy to comprehend while maintaining competitive performance against other representations. Although the TD4C algorithm has been shown to outperform the SAX representation for the task of classication of multivariate temporal data [22], it is a supervised learning method hence requires the availability of class labels. Regarding the lack of reference points for comparing similarities between time series, the use of shapelets, i.e., time series subsequences, have been proposed [23,24,25].…”
Section: Introductionmentioning
confidence: 99%
“…However, it is also possible to abstract point-based data by applying temporal knowledge which results in a more abstract representation of the data, in the form of symbolic time intervals Batal et al provide several pattern mining techniques that uses a time interval-related representation of a sequence, which requires either the events have continuous values that can be quantized or the duration of every event is available [36,42]. Moskovitch et al provide several approaches for discretizing continuous event values to derive more discriminative time-interval related patterns [40,41]. Patel et al also provide a technique for mining interval-based events [43].…”
Section: Background and Significancementioning
confidence: 99%