Mining Periodic Patterns with a MDL Criterion

Galbrun, Esther; Cellier, Peggy; Tatti, Nikolaj; Termier, Alexandre; Crémilleux, Bruno

doi:10.1007/978-3-030-10928-8_32

Cited by 13 publications

(16 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When adding a pattern to our model we search for the best possible refinement, that is we greedly extend our pattern to that version that gives us the highest gain in bits saved, we refer the reader to 'Appendix A.1' where we provide a more detailed explanation. After adding a pattern, we re-compute the residual (l. [13][14]. We repeat these steps until we have no further candidates (l. 3) and then return the final model M. In Fig.…”

Section: Algorithm 1: Omenmentioning

confidence: 99%

“…All use greedy search algorithms, iteratively adding patterns to a model until convergence. Galbrun et al [13] studied the problem of discovering sequential patterns with reliably periodically appear. Instead of taking a descriptive approach, Fowkes and Sutton [12] proposed a method to discover small sets of informative patterns based on a generative model.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Omen: discovering sequential patterns with reliable prediction delays

2022

View full text Add to dashboard Cite

Suppose we are given a discrete-valued time series $$X $$ X of observed events and an equally long binary sequence $$Y $$ Y that indicates whether something of interest happened at that particular point in time. We consider the problem of mining serial episodes, sequential patterns allowing for gaps, from $$X $$ X that reliably predict those interesting events. With reliable we mean patterns that not only predict that an interesting event is likely to follow, but in particular that we can also accurately tell how how long until that event will happen. In other words, we are specifically interested in patterns with a highly skewed distribution of delays between pattern occurrences and predicted events. As it is unlikely that a single pattern can explain a complex real-world progress, we are after the smallest, least redundant set of such patterns that together explain the interesting events well. We formally define this problem in terms of the Minimum Description Length principle, by which we identify the best patterns as those that describe the occurrences of interesting events $$Y $$ Y most succinctly given the data over $$X $$ X . As neither discovering the optimal explanation of $$Y $$ Y given a set of patterns, nor the discovery of optimal pattern set are problems that allow for straightforward optimization, we break the problem in two and propose effective heuristics for both. Through extensive empirical evaluation, we show that both our main method, Omen, and its fast approximation fOmen, work well in practice and both quantitatively and qualitatively beat the state of the art.

show abstract

Section: Algorithm 1: Omenmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Omen: discovering sequential patterns with reliable prediction delays

2022

View full text Add to dashboard Cite

show abstract

“…However, descriptive approaches using pattern mining on time series data have been mostly neglected so far, as also discussed above. An approach for compressing event logs based on the minimum description length (MDL) principle was presented by [19], making it possible to detect local patterns in temporal data. Compared to this work, which focuses on event sequences, our approach aims to find meaningful representations of (potentially complex) continuous-valued recordings, and assesses the discovered patterns by reference to a target variable rather than compression.…”

Section: Exceptionality Detection In Time Seriesmentioning

confidence: 99%

Local Exceptionality Detection in Time Series Using Subgroup Discovery

Hudson¹,

Wiltshire²,

Atzmueller³

2021

Preprint

View full text Add to dashboard Cite

In this paper, we present a novel approach for local exceptionality detection on time series data. This method provides the ability to discover interpretable patterns in the data, which can be used to understand and predict the progression of a time series. This being an exploratory approach, the results can be used to generate hypotheses about the relationships between the variables describing a specific process and its dynamics. We detail our approach in a concrete instantiation and exemplary implementation, specifically in the field of teamwork research. Using a real-world dataset of team interactions we include results from an example data analytics application of our proposed approach, showcase novel analysis options, and discuss possible implications of the results from the perspective of teamwork research.

show abstract

“…given data). The MDL principle has been successfully employed to select descriptive sets of patterns from transactional databases [13], sequence databases [5,12], relational databases [8], geometric data [4], and graphs [1].…”

Section: The MDL Principlementioning

confidence: 99%

“…Several kinds of approaches have been proposed to tackle this problem (e.g., constraint based approaches [16] or condensed representation [15]). Among them, the methods based on the Minimum Description Length (MDL) principle demonstrated that it is possible to drastically reduce the number of patterns by selecting a small set of descriptive patterns among all the generated patterns [4,5,8,12,13]. The MDL principle [6] comes from information theory, and states that the model that describes the data the best is the one that compresses the data the best, i.e.…”

Section: Introductionmentioning

confidence: 99%

GraphMDL+

Bariatti

Cellier

Ferré

2021

Proceedings of the 36th Annual ACM Symposium on Applied Computing

Self Cite

View full text Add to dashboard Cite

Graph pattern mining algorithms ease graph data analysis by extracting recurring structures. However, classic pattern mining approaches tend to extract too many patterns for human analysis. Recently, the GraphMDL algorithm has been proposed, which reduces the generated pattern set by using the Minimum Description Length (MDL) principle to select a small descriptive subset of patterns. The main drawback of this approach is that it needs to first generate all possible patterns and then sieve through their complete set.In this paper we propose GraphMDL+, an approach based on the same description length definitions as GraphMDL but which tightly interleaves pattern generation and pattern selection (instead of generating all frequent patterns beforehand), and outputs a descriptive set of patterns at any time. Experiments show that our approach takes less time to attain equivalent results to GraphMDL and can attain results that GraphMDL could not attain in feasible time. Our approach also allows for more freedom in the pattern and data shapes, since it is not tied to an external approach.

show abstract

Mining Periodic Patterns with a MDL Criterion

Cited by 13 publications

References 20 publications

Omen: discovering sequential patterns with reliable prediction delays

Omen: discovering sequential patterns with reliable prediction delays

Local Exceptionality Detection in Time Series Using Subgroup Discovery

GraphMDL+

Contact Info

Product

Resources

About