ANOMALY DETECTION TECHNIQUES AND CHALLENGES IN HIGH DIMENTIONAL DATA

The identification of anomalies in large datasets is becoming an important research topic because of the wide range of practical applications it offers. Due to so-called "big data," many current anomaly detection methods are unable to maintain acceptable accuracy. Huge dimensionality, or t...

Full description

Saved in:
Bibliographic Details
Published inNeuroQuantology Vol. 20; no. 10; p. 3220
Main Authors Iqbal, Md, Shukla, Abhishek Kumar, Jain, Anmol
Format Journal Article
LanguageEnglish
Published Bornova Izmir NeuroQuantology 01.01.2022
Subjects
Algorithms
Anomalies
Big Data
Data analysis
Datasets
Decision making
Search engines
Securities markets
Social networks
Velocity
Online AccessGet full text

Cover

More Information
Summary:The identification of anomalies in large datasets is becoming an important research topic because of the wide range of practical applications it offers. Due to so-called "big data," many current anomaly detection methods are unable to maintain acceptable accuracy. Huge dimensionality, or the "curse of big dimensionality," is a phenomenon that affects current methodologies in accordance with the performance and correctness. With high dimensionality and massive data problems, it is critical to recognise the specific concerns that arise in anomaly detection.Detecting aberrant behaviour in high-density data is a typical challenge. Inconsistencies in high-dimensional data with the property of 2V's (Volume and Velocity) should be eliminated by an effective technique. Anomaly detection in huge data is a challenging problem, yet strategies and algorithms are used, and the results of such a detection are presented and shown in this study
ISSN:1303-5150
DOI:10.14704/nq.2022.20.10.NQ55321