ROC<scp>search</scp> — An ROC-guided Search Strategy for Subgroup Discovery

Meeng, Marvin; Duivesteijn, Wouter; Knobbe, Arno

doi:10.1137/1.9781611973440.81

Cited by 8 publications

(17 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This beam only keeps the most promising subgroups to extend at each level [ [36,44,53]]. The redundancy issue due to the beam search is tackled with the pattern skyline paradigm by [54], and with a ROC-based beam search variant for SD by [42]. Another family of SD algorithms relies on evolutionary approaches.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Anytime discovery of a diverse set of patterns with Monte Carlo tree search

Bosc

Boulicaut

Raïssi

et al. 2017

Data Min Knowl Disc

View full text Add to dashboard Cite

The discovery of patterns that accurately discriminate one class label from another remains a challenging data mining task. Subgroup discovery (SD) is one of the frameworks that enables to elicit such interesting patterns from labeled data. A question remains fairly open: How to select an accurate heuristic search technique when exhaustive enumeration of the pattern space is infeasible? Existing approaches make use of beam-search, sampling, and genetic algorithms for discovering a pattern set that is nonredundant and of high quality w.r.t. a pattern quality measure. We argue that such approaches produce pattern sets that lack of diversity: Only few patterns of high quality, and different enough, are discovered. Our main contribution is then to formally define pattern mining as a game and to solve it with Monte Carlo tree search (MCTS). It can be seen as an exhaustive search guided by random simulations which can be stopped early (limited budget) by virtue of its best-first search property. We show through a comprehensive set of experiments how MCTS enables the anytime discovery of a diverse pattern set of high quality. It outperforms other approaches when dealing with a large pattern search space and for different quality measures. Thanks to its genericity, our MCTS approach can be used for SD but also for many other pattern mining tasks.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Several solutions have been proposed to filter out redundant subgroups, e.g. as did [12,53,42,10]. Basically, a neighboring function enables to keep only local optima.…”

Section: Introductionmentioning

confidence: 99%

Anytime discovery of a diverse set of patterns with Monte Carlo tree search

Bosc

Boulicaut

Raïssi

et al. 2017

Data Min Knowl Disc

View full text Add to dashboard Cite

show abstract

“…From a tool perspective, there exist several software packages for subgroup discovery, e.g., Refs . As open source options, there are, for example, subgroup discovery modules for the data mining systems Orange and RapidMiner , the Cortana system for discovering local patterns in data, as well as the specialized subgroup discovery and analytics system VIKAMINE . Using the latter a number of successful real‐world subgroup discovery applications have been implemented.…”

Section: Tools and Applicationsmentioning

confidence: 99%

“…In addition, the integration of (rich) background knowledge in a knowledge‐intensive approach, e.g., Refs , is a prerequisite for the analysis of large datasets for which relations and prior information needs to be utilized. This also tackles the area of automatic subgroup discovery, recent search strategies, e.g., Ref , and the applied significance filtering in many methods and tools . For a full‐scale approach, these issues need to be addressed such that suitable methods can be integrated comprehensively, from automatic to interactive approaches, e.g., Refs , which can also be applied for generating appropriate explanations .…”

Section: Future Directions and Challengesmentioning

confidence: 99%

Subgroup discovery

Atzmueller

2015

WIREs Data Min & Knowl

194

102

View full text Add to dashboard Cite

Subgroup discovery is a broadly applicable descriptive data mining technique for identifying interesting subgroups according to some property of interest. This article summarizes fundamentals of subgroup discovery, before that it also reviews algorithms and further advanced methodological issues. In addition, we briefly discuss tools and applications of subgroup discovery approaches. In that context, we also discuss experiences and lessons learned and outline some of the future directions in order to show the advantages and benefits of subgroup discovery.

show abstract

“…When the features serve as onedimensional subgroups on the first level of the search lattice, the entire realizations of A are used. For one-dimensional refinements, only those realizations covered by the subgroup in consideration are used [13,20]. For readability, we keep our discussion to the first case.…”

Section: Mining Binary Featuresmentioning

confidence: 99%

Flexibly Mining Better Subgroups

Nguyen

Vreeken

2016

Proceedings of the 2016 SIAM International Conference on Data Mining

View full text Add to dashboard Cite

Finding patterns from binary data is a classical problem in data mining, dating back to at least frequent itemset mining. More recently, approaches such as tiling and Boolean matrix factorization (BMF), have been proposed to find sets of patterns that aim to explain the full data well. These methods, however, are not robust against non-trivial destructive noise, i.e. when relatively many 1s are removed from the data: tiling can only model additive noise while BMF assumes approximately equal amounts of additive and destructive noise. Most real-world binary datasets, however, exhibit mostly destructive noise. In presence/absence data, for instance, it is much more common to fail to observe something than it is to observe a spurious presence. To address this problem, we take the recent approach of employing the Minimum Description Length (MDL) principle for BMF and introduce a new algorithm, Nassau, that directly optimizes the description length of the factorization instead of the reconstruction error. In addition, unlike the previous algorithms, it can adjust the factors it has discovered during its search. Empirical evaluation on synthetic data shows that Nassau excels at datasets with high destructive noise levels and its performance on real-world datasets confirms our hypothesis of the high numbers of missing observations in the real-world data

show abstract

ROCsearch — An ROC-guided Search Strategy for Subgroup Discovery

Cited by 8 publications

References 16 publications

Anytime discovery of a diverse set of patterns with Monte Carlo tree search

Anytime discovery of a diverse set of patterns with Monte Carlo tree search

Subgroup discovery

Flexibly Mining Better Subgroups

Contact Info

Product

Resources

About