A Relaxation-Based Approach for Mining Diverse Closed Patterns

Hien, Arnold; Loudni, Samir; Aribi, Noureddine; Lebbah, Yahia; Laghzaoui, Mohammed El Amine; Ouali, Abdelkader; Zimmermann, Albrecht

doi:10.1007/978-3-030-67658-2_3

Cited by 7 publications

(14 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [15], we showed that the Jaccard index has no monotonicity property, which prevents usual pruning techniques and makes classical pattern mining unworkable. To cope with the non-monotonicity of the Jaccard similarity, we proposed an anti-monotonic lower bound relaxation of the Jaccard index, which allow effective pruning.…”

Section: Pattern Diversitymentioning

confidence: 99%

“…By using this lower bound, it is easier to filter items that would give patterns too close to the ones in the history (see [15] for more details).…”

Section: Pattern Diversitymentioning

confidence: 99%

“…The filtering of cdFlexics combines two propagators, that of closedDiversity (see [15] for more details) and the propagator for XOR constraints. In CP, constraints interact through shared variables, i.e., as soon as a constraint has been propagated (no more values can be pruned), and at least a value has been eliminated from the domain of a variable, say v, then the propagation algorithms of all other constraints involving v are triggered.…”

Section: Filtering Of Cdflexicsmentioning

confidence: 99%

“…Its starts by calculating the ICV of all items of the patterns p ∈ P (loop 5−10). Then, it iteratively combines the items to form a larger and finer-grained discriminating pattern (loop [11][12][13][14][15][16][17][18][19] . Obviously, before combining sub-itemsets, we should ensure that the resulting sub-pattern belongs to an existing pattern p ∈ P (line 12).…”

Section: Learning Discriminating Sub-patternsmentioning

confidence: 99%

“…Constraint-based pattern mining has been leveraged as a general, flexible solution to pattern mining for a while [28][29][30], and has since seen numerous improvements, including closed itemset mining [31] and, eventually mining of diverse itemsets [15].…”

Section: Evaluating Aspects Of Letsip-disc-cdfmentioning

confidence: 99%

See 4 more Smart Citations

Exploiting complex pattern features for interactive pattern mining

Hien¹,

Loudni²,

Aribi³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Recent years have seen a shift from a pattern mining process that has users define constraints before-hand, and sift through the results afterwards, to an interactive one. This new framework depends on exploiting user feedback to learn a quality function for patterns. Existing approaches have a weakness in that they use static pre-defined low-level features, and attempt to learn independent weights representing their importance to the user. As an alternative, we propose to work with more complex features that are derived directly from the pattern ranking imposed by the user. Learned weights are then aggregated onto lower-level features and help to drive the quality function in the right direction. We explore the effect of different parameter choices experimentally and find that using higher-complexity features leads to the selection of patterns that are better aligned with a hidden quality function while not adding significantly to the run times of the method. Getting good user feedback requires to quickly present diverse patterns, something that we achieve but pushing an existing diversity constraint into the sampling component of the

show abstract

Section: Pattern Diversitymentioning

confidence: 99%

“…By using this lower bound, it is easier to filter items that would give patterns too close to the ones in the history (see [15] for more details).…”

Section: Pattern Diversitymentioning

confidence: 99%

Section: Filtering Of Cdflexicsmentioning

confidence: 99%

Section: Learning Discriminating Sub-patternsmentioning

confidence: 99%

Section: Evaluating Aspects Of Letsip-disc-cdfmentioning

confidence: 99%

See 3 more Smart Citations

Exploiting complex pattern features for interactive pattern mining

Hien¹,

Loudni²,

Aribi³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Associative Classification Based on the Table Constraint Satisfaction

Zuenko¹

2023

Studies in Systems, Decision and Control

View full text Add to dashboard Cite

Interactive Pattern Mining Using Discriminant Sub-patterns as Dynamic Features

Hien

Loudni

Aribi

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Recent years have seen a shift from a pattern mining process that has users define constraints before-hand, and sift through the results afterwards, to an interactive one. This new framework depends on exploiting user feedback to learn a quality function for patterns. Existing approaches have a weakness in that they use static pre-defined low-level features, and attempt to learn independent weights representing their importance to the user. As an alternative, we propose to work with more complex features that are derived directly from the pattern ranking imposed by the user. Those features are used to learn weights to be aggregated with low-level features and help to drive the quality function in the right direction. Experiments on UCI datasets show that using higher-complexity features leads to the selection of patterns that are better aligned with a hidden quality function while being competitively fast when compared to state-of-the-art methods.

show abstract

A Relaxation-Based Approach for Mining Diverse Closed Patterns

Cited by 7 publications

References 18 publications

Exploiting complex pattern features for interactive pattern mining

Exploiting complex pattern features for interactive pattern mining

Associative Classification Based on the Table Constraint Satisfaction

Interactive Pattern Mining Using Discriminant Sub-patterns as Dynamic Features

Contact Info

Product

Resources

About