Exact Primitives for Time Series Data Mining

Mueen, Sbdullah Al

doi:10.1145/2339530.2378374

Cited by 2 publications

(2 citation statements)

References 87 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because of the multi-dimensionality, complexity, indeterminacy and dynamic character in practical time series, DM & KD of time series is known as one of the ten challenging problems in data mining research [1] . Contents of DM & KD of time series are far-ranging, such as bursts, periods, motifs, outliers and shapelets, etc [2] , which are very useful in many different fields such as culture I-Ching [3] , signal processing of spacecraft, navigation and guidance, fault diagnosis, prognostics & health management (PHM), automation, iatrology, biology and economics etc. Up to now, most of researches are paid attention to outliers mining, pattern recognition, mode analysis, trend prediction for one-dimensional time series and statistical character distilling for multi-dimensional stationary time series [4][5][6][7] .…”

Section: Introductionmentioning

confidence: 99%

Algorithms of Data Mining and Knowledge Discovery of Correlativity in Two-Dimensional Time Series

Zhang

2012

AMM

View full text Add to dashboard Cite

Oriented at dynamic data from complicated process with noise disturbance, it is very difficult to discover knowledge of correlativity and orderliness. Following some analyzing results about the shortcoming of relative coefficients in mining non-stationary time series, a series of new algorithms are built in this paper to mine correlativity in two-dimensional time series. These new algorithms are based on a expansible framework of model set. Based on these new mining algorithms, a making decision table is listed not only to mine correlativity in two-dimensional time series, but also to discover deepening knowledge to transform the qualitative knowledge “nonlinear relativity” as well as “non-relativity” into deeper quantitative knowledge about analytical model. These new approaches given in this paper is exoteric in framework and can be enriched with additional new models. In this way, some professional data mining and knowledge discovery cab be fulfilled to aim at some specific professional fields.

show abstract

Section: Introductionmentioning

confidence: 99%

Algorithms of Data Mining and Knowledge Discovery of Correlativity in Two-Dimensional Time Series

Zhang

2012

AMM

View full text Add to dashboard Cite

show abstract

“…generates all the possible pairs. Lemma 3.2 states exactness of the algorithm [22] as it ensures search operation to be performed in all possible pairs. Interested readers may find proofs to the lemmas in the original literature [22].…”

Section:  mentioning

confidence: 99%

Time Series Pattern Discovery Techniques for Well-to-Well Log Correlation

Apiwatsakulchai

View full text Add to dashboard Cite

Well log data is one of the most abundant data sources for subsurface characterization since almost all drilled wells are logged. Subsurface characterization requires correlated well log information. Processes to help perform well-to-well log correlation have been developed in this study based on the two approaches: pattern matching and pattern discovery. Pattern matching focuses on finding a known pattern of interest in uncorrelated wells. The use of Euclidean, Hamming, and Levenshtein distances in similarity measurement, Piecewise Aggregate Approximation (PAA), and Symbolic Aggregate approXimation (SAX) data representation helps create reduced-resolution while maintains the main characteristics of the signals. Multi-resolution analysis provides the most probable match from different data resolutions obtained by different discretization levels. The results show that multi-resolution analysis can identify the most probable match from the most frequent data window promoted. The results also show that Levenshtein distance is far superior to Hamming and Euclidean distances in finding the best match even when local variations are present. Pattern discovery focuses on finding repeating patterns without any prior knowledge of patterns that might exist in the well logs. MK (Motif Kymatology) is used as the base algorithm in this study. For the data sets tried in this study, the proposed algorithm can successfully identify repeating patterns. A modified step is incorporated in order to help increase stability of the final correlated well sections. In summary, pattern matching and pattern discovery can be formed as an integrated workflow to support well log correlation task and subsurface characterization.

show abstract

Exact Primitives for Time Series Data Mining

Cited by 2 publications

References 87 publications

Algorithms of Data Mining and Knowledge Discovery of Correlativity in Two-Dimensional Time Series

Algorithms of Data Mining and Knowledge Discovery of Correlativity in Two-Dimensional Time Series

Time Series Pattern Discovery Techniques for Well-to-Well Log Correlation

Contact Info

Product

Resources

About