Comparison of Machine Learning and Statistical Approaches of Detecting Anomalies Using a Simulation Study - Lower Silesian Digital Library

Skip to main menu
Skip to search engine
Skip to content
Skip to footer

Comparison of Machine Learning and Statistical Approaches of Detecting Anomalies Using a Simulation Study

Download

Title: Comparison of Machine Learning and Statistical Approaches of Detecting Anomalies Using a Simulation Study

Title in english:

Uczenie maszynowe i statystyczne metody wykrywania anomalii – porównawcza analiza symulacyjna

Creator:

Lenart, Klaudia

Description:

Econometrics = Ekonometria, 2024, Vol. 28, No. 4, s. 23-31

Abstrakt:

Aim: An anomaly is an observation or a group of observations that is unusual for a given dataset. Anomaly detection has many applications, not only as a step of data preparation but also, for example, as a way of identifying credit card fraud detection, network intrusions and much more. There are diverse methods of anomaly detection. In particular two groups of methods have been developed independently – statistical methods and machine learning algorithms. Those methods are rarely compared. While statistical methods focus on formulating a measure of the abnormality of the observations, supervised machine learning makes it possible to use data about typical observations and previously identified anomalies. The aim of this paper was to compare the two approaches by conducting a simulation study. Methodology: A simulation study was conducted, during which the data was generated using copula functions. For the purpose of generating different types of anomalies, marginal distributions of the variables were manipulated. The effectiveness of each method was evaluated based on measures of classification model performance. Results: While the accuracy of the statistical methods was dependent on the precise prediction of the percentage of the anomalies that would occur in the data, the machine learning algorithms’ recall was significantly lower when the change in the marginal distribution of the value parameters was smaller. Implications and recommendations: For the statistical methods included in the study, knowledge about the distribution of the variables was crucial while the supervised machine learning algorithms required acquiring a training dataset. Unlike machine learning algorithms, the statistical methods performed with similar accuracy even when the change in the marginal distribution parameters’ value was smaller Originality/value: The two approaches to anomaly detection presented in the paper are not often compared, usually used by two separate groups of researchers – statisticians and machine learning or data science specialists.

Publisher:

Publishing House of Wroclaw University of Economics and Business

Place of publication:

Date:

Resource Type:

Resource Identifier:

doi:10.15611/eada.2024.4.02 ; oai:dbc.wroc.pl:131964

Language:

Relation:

Econometrics = Ekonometria, 2024, Vol. 28, No. 4

Rights:

Pewne prawa zastrzeżone na rzecz Autorów i Wydawcy

Access Rights:

Dla wszystkich zgodnie z licencją

License:

CC BY-SA 4.0

Location:

Uniwersytet Ekonomiczny we Wrocławiu

Group publication title:

Ekonometria = Econometrics

Subject and keywords:

anomaly detection simulation study machine learning wykrywanie anomalii badanie symulacyjne uczenie maszynowe

Object collections:

Last modified:

Feb 27, 2025

In our library since:

Feb 27, 2025

Number of object content hits:

5

All available object's versions:

https://dbc.wroc.pl./publication/170707

Show description in RDF format:

Show description in OAI-PMH format:

Ekonometria = Econometrics

Edition name	Date
Comparison of Machine Learning and Statistical Approaches of Detecting Anomalies Using a Simulation Study	Feb 27, 2025

Objects Similar

Informatyka Ekonomiczna = Business Informatics

Informatyka Ekonomiczna = Business Informatics

Attack vectors on supervised machine learning systems in business applications

Surma, Jerzy

2020

artykuł

Prace Naukowe Uniwersytetu Ekonomicznego we Wrocławiu

Prace Naukowe Uniwersytetu Ekonomicznego we Wrocławiu

Artificial Intelligence in audit

Karmańska, Anna

2022

artykuł

Debiuty Studenckie

Debiuty Studenckie

Budowa i testowanie wsteczne strategii inwestycyjnej opartej o uczenie maszynowe

Wietrzykowski, Filip Grześkowiak, Alicja. Redakcja Peternek, Piotr. Redakcja

2023

rozdział

Debiuty Studenckie

Debiuty Studenckie

Diagnoza nierówności płacowych wśród sportowców z wykorzystaniem metod uczenia maszynowego

Derda, Aleksandra Stanimir, Agnieszka. Redakcja

2024

rozdział

Ekonomia XXI Wieku = Economics of the 21st Century

Ekonomia XXI Wieku = Economics of the 21st Century

Wykorzystanie Google Trends do predykcji stopy zwrotu indeksu WIG20

Niedzielska, Ewelina

2018

artykuł

Informatyka Ekonomiczna = Business Informatics

Informatyka Ekonomiczna = Business Informatics

Chatbot – modelowanie i aplikacje

Szymański, Grzegorz Jóźwiak, Piotr

2018

artykuł

Debiuty Studenckie

Debiuty Studenckie

Fairness in machine learning – bias identification and reduction

Zaniewska, Aleksandra Dudycz, Helena. Redaktor

2022

rozdział

Prace Naukowe Uniwersytetu Ekonomicznego we Wrocławiu

Prace Naukowe Uniwersytetu Ekonomicznego we Wrocławiu

Skuteczność metod klasyfikacji w prognozowaniu kierunku zmian indeksu giełdowego S&P500

Krawczuk, Jerzy

2012

artykuł

More

Citation style:

Chicago ISO690 APA IEEE Harvard

Contact details

Address

Wrocław University of Science and Technology (Coordinator)
Library of Wrocław University of Science and Technology
Wybrzeże Wyspiańskiego 27
50-370 Wrocław, Poland

Phone

(+48) 71-320-38-30
(+48) 71-320-47-50

E-Mail

Visit us!

https://biblioteka.pwr.edu.pl/

This page uses 'cookies'. More information