A Method for Temporal Knowledge Conversion

Guimarães, Gabriela; Ultsch, Alfred

doi:10.1007/3-540-48412-4_31

Cited by 21 publications

(14 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Symbolic interval time series are an important data format for discovering temporal knowledge that can be easily communicated to human analysts (Guimarães and Ultsch 1999;Kam and Fu 2000;Villafane et al 2000;Höppner 2001Höppner , 2003Cohen 2001;Last et al 2001;Mörchen et al 2004;Bellazi et al 2005;Papaterou et al 2005;Winarko and Roddick 2007). Numerical time series are often converted to symbolic interval time series by segmentation (Last et al 2001;Höppner 2003), discretization (Villafane et al 2000;Mörchen et al 2004;Bellazi et al 2005) or clustering (Guimarães and Ultsch 1999;Mörchen et al 2004) Alternatively, interval data can be obtained directly from other temporal data, e.g., video (Fern 2004) or association rules over time (Rainsford and Roddick 1999). Patterns mined from symbolic interval data can provide explanation for the underlying temporal processes or anomalous behavior.…”

Section: Introductionmentioning

confidence: 98%

Efficient mining of understandable patterns from multivariate interval time series

Mörchen

Ultsch

2007

Data Min Knowl Disc

Self Cite

View full text Add to dashboard Cite

We present a new method for the understandable description of local temporal relationships in multivariate data, called Time Series Knowledge Mining (TSKM). We define the Time Series Knowledge Representation (TSKR) as a new language for expressing temporal knowledge in time interval data. The patterns have a hierarchical structure, with levels corresponding to the temporal concepts duration, coincidence, and partial order. The patterns are very compact, but offer details for each element on demand. In comparison with related approaches, the TSKR is shown to have advantages in robustness, expressivity, and comprehensibility. The search for coincidence and partial order in interval data can be formulated as instances of the well known frequent itemset problem. Efficient algorithms for the discovery of the patterns are adapted accordingly. A novel form of search space pruning effectively reduces the size of the mining result to ease interpretation and speed up the algorithms. Human interaction is used during the mining to analyze and validate partial results as early as possible and guide further processing steps. The efficacy of the methods is demonstrated using two real life data sets. In an application to sports medicine the results were recognized as valid and useful by an expert of the field.

show abstract

Section: Introductionmentioning

confidence: 98%

Efficient mining of understandable patterns from multivariate interval time series

Mörchen

Ultsch

2007

Data Min Knowl Disc

Self Cite

View full text Add to dashboard Cite

show abstract

“…Ultsch had proposed the UTG [9], [10], which is a hierarchical pattern language for the temporal knowledge discovery, and he had also implemented it [4]. Temporal abstraction is increased through each level of the hierarchy.…”

Section: Related Work and Motivationmentioning

confidence: 99%

Procedural Steps for Knowledge Mining in Time Series

ChandraDev¹,

Behera²

2013

IJCA

View full text Add to dashboard Cite

Symbolic intervals which form temporal patterns are usually formulated through Allen's interval relations that originate in temporal reasoning. But this representation is not advantages for knowledge discovery. The Hierarchical Time series Knowledge Representation (HTKR) is the hierarchical language which expresses the temporal aspects of coincidence and partial order, for interval patterns. We present mining procedural steps which are more efficient, effective and based on item set techniques. Pruning of the search space minimizes the mining result size considerably, thereby speeding up the procedural steps and easing the interpretations. When applied on the real data set, HTKR can provide the explanation of underlying temporal phenomena, but whereas the numerous Allen's relation patterns only explains fragmented data. General TermsAlgorithms.

show abstract

“…Due to lack of space we will only briefly describe the conceptual levels of the hierarchy (see also Figure 1) along with an example from the application. The basic ideas of the UTG were developed in Ultsch (1996) and applied in Guimaraes and Ultsch (1999). For a detailed description see Ultsch (2004).…”

Section: Unification-based Temporal Grammarmentioning

confidence: 99%

“…If the process alternates between several regimes or states, these regions should form clusters in the high dimensional space obtained disregarding the time attribute. In Guimaraes and Ultsch (1999) and for the identification of the skating movement phases Emergent Self-Organizing Maps (ESOM) ) have been used to identify clusters. The rules for each cluster were generated using the Sig * Algorithm (Ultsch (1991)).…”

Section: Time Series Knowledge Miningmentioning

confidence: 99%

See 1 more Smart Citation

Discovering Temporal Knowledge in Multivariate Time Series

Mörchen

Ultsch

Studies in Classification, Data Analysis, and Knowledge Organization

Self Cite

View full text Add to dashboard Cite

Abstract. An overview of the Time Series Knowledge Mining framework to discover knowledge in multivariate time series is given. A hierarchy of temporal patterns, which are not a priori given, is discovered. The patterns are based on the rule language Unification-based Temporal Grammar. A semiotic hierarchy of temporal concepts is build in a bottom up manner from multivariate time instants. We describe the mining problem for each rule discovery step. Several of the steps can be performed with well known data mining algorithms. We present novel algorithms that perform two steps not covered by existing methods. First results on a dataset describing muscle activity during sports are presented.

show abstract

A Method for Temporal Knowledge Conversion

Cited by 21 publications

References 6 publications

Efficient mining of understandable patterns from multivariate interval time series

Efficient mining of understandable patterns from multivariate interval time series

Procedural Steps for Knowledge Mining in Time Series

Discovering Temporal Knowledge in Multivariate Time Series

Contact Info

Product

Resources

About