Discovering Temporal Knowledge in Multivariate Time Series

Mörchen, Fabian; Ultsch, Alfred

doi:10.1007/3-540-28084-7_30

Cited by 16 publications

(13 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The time series Knowledge Discovery framework Temporal Data Mining Method (TDM) [1] is a framework of methods and algorithms to mine rules in multivariate time series (MVTS). The patterns are expressed with the hierarchical temporal rule language Unification-based Temporal Grammar (UTG) [2].…”

Section: Temporal Data Mining Methodsmentioning

confidence: 99%

Discovering Interpretable Muscle Activation Patterns with the Temporal Data Mining Method

Mörchen

Ultsch

Hoos

2004

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. The understanding of complex muscle coordination is an important goal in human movement science. There are numerous applications in medicine, sports, and robotics. The coordination process can be studied by observing complex, often cyclic movements, which are dynamically repeated in an almost identical manner. In this paper we demonstrate how interpretable temporal patterns can be discovered within raw EMG measurements collected from tests in professional In-Line Speed Skating. We show how the Temporal Data Mining Method, a general framework to discover knowledge in multivariate time series, can be used to extract such temporal patterns. This representation of complex muscle coordination opens up new possibilities to optimize, manipulate, or imitate the movements.

show abstract

Section: Temporal Data Mining Methodsmentioning

confidence: 99%

Discovering Interpretable Muscle Activation Patterns with the Temporal Data Mining Method

Mörchen

Ultsch

Hoos

2004

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…Each segment except the first and last segments has two sets of clusters, one set dissimilar to the clusters in previous window and the other set dissimilar to the clusters in the next window. We are interested in the genes that are significantly clustered together in these two sets of clusters, as they segment [1][2][3][4][5][6] segment [7][8][9][10] segment [11][12][13][14] represent the genes that are specific to the segment under consideration. We calculate a contingency table of these two clusterings for each segment (excluding the first and the last segment).…”

Section: B Dataset Segmentation and Kripke Model State Inferencementioning

confidence: 99%

“…Algorithms to mine the temporal order of events occurring in multiple time series have also been well studied. Moerchen et al [7] devised a temporal grammar for this purpose. However, their approach requires manual partitioning of the time series, and the events are derived by naive discretization of the multiple time series.…”

Section: Introductionmentioning

confidence: 99%

Deriving Kripke structures from time series segmentation results

Tadepalli

Ramakrishnan

Mishra

et al. 2008

2008 9th International Workshop on Discrete Event Systems

View full text Add to dashboard Cite

Abstract-Kripke structures are important modeling formalisms to understand the behavior of reactive systems. We present an approach to automatically infer Kripke structures from time series datasets. Our algorithm bridges the continuous world of time profiles and the discrete symbols of Kripke structures by incorporating a segmentation algorithm as an intermediate step. This approach identifies, in an unsupervised manner, the states of the Kripke structure, the transition relation, and the properties (propositions) that hold true in each state. We demonstrate experimental results of our approach to understanding the interplay between key biological processes.

show abstract

“…4,5 Other works extract common or recurrent patterns in multivariate time series, describing that several time series have exactly the same profile. 4,6 However, to the best of our knowledge, there does not exist any data mining method that allows the discovery of frequent co-evolutions, not necessarily following the same trends, in a sequence database.…”

Section: Introductionmentioning

confidence: 99%

Evolution patterns and gradual trends

Fiot

Masséglia

Laurent

et al. 2009

Int. J. Intell. Syst.

View full text Add to dashboard Cite

Nowadays, many databases record ordered or temporally annotated data, such as Web access logs or genomic sequences. Therefore, sequence mining has become an important research area. Among these data mining approaches, sequential patterns aim at describing frequent behaviors. In the access data of a commercial Web site, one may, for instance, discover that "35% of customers successively buy a PSP then a memory stick and PSP games". To provide more complete information, fuzzy sequential patterns were designed, including quantitative values within the mining task. Such patterns, considering the previous example, would be "35% of customers buy a PSP, then they buy few games many times, and then they buy a high-capacity memory stick once." However, symbolic or fuzzy sequential patterns, in their current form, do not allow to extract temporal tendencies that are typical of sequential data. By means of temporal tendency mining, one may discover in the same access data that "An increasing number of purchases of PSP games during a very short period is frequently followed by a purchase of a high-capacity memory stick a few days later." It would be easy to conclude that the users either quickly succeed in registering or make several attempts before they look at the help page within a few seconds. To the best of our knowledge, no method has been designed for discovering this kind of patterns. Therefore, we propose, in this paper, two approaches that extract pattern-expressing trends or evolution. First, we define evolution patterns that summarize the evolution of the quantities in the data. We explain how they can be obtained from a quantitative sequence database. Second, we define gradual trends in fuzzy sequential data. These trends describe variations in the fulfillment of fuzzy properties according to time. For both kinds of patterns, we developed algorithms that were implemented and tested on real data. C 2009 Wiley Periodicals, Inc.

show abstract

Discovering Temporal Knowledge in Multivariate Time Series

Cited by 16 publications

References 10 publications

Discovering Interpretable Muscle Activation Patterns with the Temporal Data Mining Method

Discovering Interpretable Muscle Activation Patterns with the Temporal Data Mining Method

Deriving Kripke structures from time series segmentation results

Evolution patterns and gradual trends

Contact Info

Product

Resources

About