FSSD - A Fast and Efficient Algorithm for Subgroup Set Discovery

Belfodil, Adnene; Belfodil, Aimene; Bendimerad, Anes; Lamarre, Philippe; Robardet, Céline; Kaytoue, Mehdi; Plantevit, Marc

doi:10.1109/dsaa.2019.00023

Cited by 15 publications

(14 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, although we consider a rich pattern language in comparison to other approaches, we can exploit the syntactic tree structure of the queries, to mine tree patterns, more expressive than conjunctions of SQL clauses. Third, an effort is needed to produce more qualitative subgroup sets with more diversity (data cover) and less redundancy [44], and directly considering a quality measure on the subgroup set, turning the top-k mining problem into subgroup set mining [45]. Finally, subjectivness and practitioner preferences should be considered by the mining algorithms through an interactive discovery process.…”

Section: Discussionmentioning

confidence: 99%

"What makes my queries slow?": Subgroup Discovery for SQL Workload Analysis

Remil¹,

Bendimerad²,

Mathonat³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Among daily tasks of database administrators (DBAs), the analysis of query workloads to identify schema issues and improving performances is crucial. Although DBAs can easily pinpoint queries repeatedly causing performance issues, it remains challenging to automatically identify subsets of queries that share some properties only (a pattern) and simultaneously foster some target measures, such as execution time. Patterns are defined on combinations of query clauses, environment variables, database alerts and metrics and help answer questions like what makes SQL queries slow? What makes I/O communications high? Automatically discovering these patterns in a huge search space and providing them as hypotheses for helping to localize issues and root-causes is important in the context of explainable AI. To tackle it, we introduce an original approach rooted on Subgroup Discovery. We show how to instantiate and develop this generic data-mining framework to identify potential causes of SQL workloads issues. We believe that such data-mining technique is not trivial to apply for DBAs. As such, we also provide a visualization tool for interactive knowledge discovery. We analyse a one week workload from hundreds of databases from our company, make both the dataset and source code available, and experimentally show that insightful hypotheses can be discovered.

show abstract

Section: Discussionmentioning

confidence: 99%

"What makes my queries slow?": Subgroup Discovery for SQL Workload Analysis

Remil¹,

Bendimerad²,

Mathonat³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…It has been shown that Problem the non-redundant subgroup set discovery problem is equivalent to the discovery of closed on the positive patterns (COTP): [12] addresses the general case while [7], [11] consider the discovery of closed on the positive interval patterns within numerical data. The key idea is that any sub-interval of a COTP will drop at least one positive object, thus, reducing the true positive rate, hence many measures such as the W RAcc.…”

Section: Closed On the Positive Interval Patternsmentioning

confidence: 99%

Anytime Subgroup Discovery in High Dimensional Numerical Data

Mathonat¹,

Nurbakova

Boulicaut

et al. 2021

2021 IEEE 8th International Conference on Data Science and Advanced Analytics (DSAA)

Self Cite

View full text Add to dashboard Cite

Subgroup discovery (SD) enables one to elicit patterns that strongly discriminate a class label. When it comes to numerical data, most of the existing SD approaches perform data discretizations and thus suffer from information loss. A few algorithms avoid such a loss by considering the search space of every interval pattern built on the dataset numerical values and provide an "anytime" property: at any moment, they are able to provide a result that improves over time. Given a sufficient time/memory budget, they may eventually complete an exhaustive search. However, such approaches are often intractable when dealing with high-dimensional numerical data, for instance, when extracting features from real-life multivariate time series. To overcome such limitations, we propose MonteCloPi, an approach based on a bottom-up exploration of numerical patterns with a Monte Carlo Tree Search. It enables to have a better exploration-exploitation trade-off between exploration and exploitation when sampling huge search spaces. Our extensive set of experiments proves the efficiency of MonteCloPi on highdimensional data with hundreds of attributes. We finally discuss the actionability of discovered subgroups when looking for skill analysis from Rocket League action logs.

show abstract

“…e training results of this algorithm are compared with the traditional SSD, RFBNet (Receptive Field Block Net) [33], FSSD (Feature Fusion Single-Shot Multibox Detector) [34], RefineDet (Single-Shot Refinement Neural Network for Object Detection) [35], and M2Det (Multilevel and Multiscale Detector) [36] algorithm, where the x-axis represents the period (epoch) and the y-axis represents the training loss. In the early stage of training, the RefineDet uses the Refi-ne_multibox_loss, which is the second operation of the multibox_loss, so the convergence speed is slow; the other 5 curves use the multibox_loss as the loss function, and their convergence speeds are very fast.…”

Section: Rationality Analysis Of Feature Cross-level Fusionmentioning

confidence: 99%

A Marine Object Detection Algorithm Based on SSD and Feature Enhancement

Lu²,

Lu³

et al. 2020

Complexity

View full text Add to dashboard Cite

Autonomous detection and fishing by underwater robots will be the main way to obtain aquatic products in the future; sea urchins are the main research object of aquatic product detection. When the classical Single-Shot MultiBox Detector (SSD) algorithm is applied to the detection of sea urchins, it also has disadvantages of being inaccurate to small targets and insensitive to the direction of the sea urchin. Based on the classic SSD algorithm, this paper proposes a feature-enhanced sea urchin detection algorithm. Firstly, according to the spiny-edge characteristics of a sea urchin, a multidirectional edge detection algorithm is proposed to enhance the feature, which is taken as the 4th channel of image and the original 3 channels of underwater image together as the input for the further deep learning. Then, in order to improve the shortcomings of SSD algorithm’s poor ability to detect small targets, resnet 50 is used as the basic framework of the network, and the idea of feature cross-level fusion is adopted to improve the feature expression ability and strengthen semantic information. The open data set provided by the National Natural Science Foundation of China underwater Robot Competition will be used as the test set and training set. Under the same training and test conditions, the AP value of the algorithm in this paper reaches 81.0%, 7.6% higher than the classic SSD algorithm, and the confidence of small target analysis is also improved. Experimental results show that the algorithm in this paper can effectively improve the accuracy of sea urchin detection.

show abstract

FSSD - A Fast and Efficient Algorithm for Subgroup Set Discovery

Cited by 15 publications

References 44 publications

"What makes my queries slow?": Subgroup Discovery for SQL Workload Analysis

"What makes my queries slow?": Subgroup Discovery for SQL Workload Analysis

Anytime Subgroup Discovery in High Dimensional Numerical Data

A Marine Object Detection Algorithm Based on SSD and Feature Enhancement

Contact Info

Product

Resources

About