Efficient mining of understandable patterns from multivariate interval time series

Abstract

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.

Mining Signatures from Event Sequences

“…There is considerably less work in the area of sequence retrieval, and the problem is more general and difficult. For more detail about time series and sequence retrieval can be found in Das and Gunopulos (2003) [15].…”

“…For instance, Line al. [12] For information representation of discrete time series data, Moerchen et al [14], [16], [15] proposed a novel Time Series Knowledge Representation (TSKR) as a pattern language (grammar) for temporal knowledge discovery from multivariate time series and symbolic interval data, where the secular knowledge representation is in the form of symbolic languages and grammars that have been formulated as a means to perform intelligent reasoning and inference from time-dependent event orders.…”

“…The first set of data is constructed to test the effectiveness of our individual OSC-NMF approach, which made of Moerchens's TSKR event-interval-test-pattern [14], [15], [16]. that has been abstracted from a tutorial figure.…”

Effective elimination of redundant association rules

“…This concept has been successfully applied to query graph databases (Cheng et al, 2007a) and other possible applications include redundancy elimination in mining time series data (Mörchen and Ultsch, 2007) and data stream mining (Cheng et al, 2007b). This paper applies the concept to effectively remove the redundancy in the set of ARs.…”

Time series data mining using complex networks

“…For example, considering that the price of a stock A 1 is increasing and that A 2 is decreasing, then there is x% of the stock price A 3 also decreasing. Several works have been developed in this area (CHEN;PETROUNIAS, 2000;LEE;CHEN, 2001;SARKER et al, 2003;MöRCHEN;ULTSCH, 2007). …”