Note on the curse of dimensionality regarding financial anomaly detection
by Sarah Oliveira Pinto; Vinicius Amorim Sobreiro
International Journal of Business Intelligence and Systems Engineering (IJBISE), Vol. 2, No. 1, 2024

Abstract: The so-called curse of dimensionality represents limitations to the processes of manipulation, interpretation and analysis of high dimensional data, which makes identifying behaviour patterns in cursed dimensional feature spaces complex. Several recently published articles discussed the need to use dimensionality reduction methods when considering the issue of detecting anomalies in financial systems. However, most articles developed models that did not consider real-time data or the continuous generation of information. They also did not directly address how to overcome situations arising from a great volume of data. This short communication reports on articles that used voluminous but static data, and articles that used real-time data to identify anomalies in financial systems but did not significantly address the issues related to dimensionality. Furthermore, this note signals the need for further investigation of how to break the curse of dimensionality through developing financial anomaly detection models to promote better decision support systems.

Online publication date: Sat, 15-Jun-2024

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Business Intelligence and Systems Engineering (IJBISE):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com