Reducing distance computations for distance-based outliers

•The paper introduces the FastSolvingSet algorithm to discover outliers.•This algorithm computes the distance based outliers with no approximation.•The experiments outline that a large amount of distance computations is saved.•FastSolvingSet is suitable to be used in parallel/distributed scenarios....

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Expert systems with applications Ročník 147; s. 113215
Hlavní autoři: Angiulli, Fabrizio, Basta, Stefano, Lodi, Stefano, Sartori, Claudio
Médium: Journal Article
Jazyk:angličtina
Vydáno: New York Elsevier Ltd 01.06.2020
Elsevier BV
Témata:
Algorithms
Data analysis
Distance-based outliers
Fraud
Intrusion
Molecular biology
Optimization
Outlier detection
Outliers (statistics)
Parallel and distributed algorithms
Parallel and distributed algorithms
Outlier detection
Distance-based outliers
ISSN:0957-4174, 1873-6793
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:•The paper introduces the FastSolvingSet algorithm to discover outliers.•This algorithm computes the distance based outliers with no approximation.•The experiments outline that a large amount of distance computations is saved.•FastSolvingSet is suitable to be used in parallel/distributed scenarios. The mining task of outlier detection is essential in many expert and intelligent systems exploited in a wide range of applications, from intrusion detection to molecular biology. In some of such applications the ability to process large amounts of data in a very short time can be critical, for instance in intrusion and fraud detection. This paper explores a solution for the optimisation of an exact, unsupervised outlier detection method by avoiding unnecessary computations, and therefore reducing the running time and making the method usable also in settings where response times are crucial. In particular, we enhance the SolvingSet-based approach by using a mechanism that exploits the knowledge learned during the algorithm execution and avoids a large amount of distance computations. We demonstrate the strength of the proposed solution, named FastSolvingSet, through both theoretical and experimental analysis.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0957-4174
1873-6793
DOI:10.1016/j.eswa.2020.113215