Welcome to the InfoSci Platform
Simulated Workbench Design for Characterisation and Selection of Appropriate Outlier Ensemble Algorithm

Simulated Workbench Design for Characterisation and Selection of Appropriate Outlier Ensemble Algorithm

Divya D., M. Bhasi, Santosh Kumar
Copyright: © 2022 |Volume: 13 |Issue: 1 |Pages: 23
ISSN: 1947-8186|EISSN: 1947-8194|EISBN13: 9781683181699|DOI: 10.4018/IJISMD.315024
Cite Article Cite Article

MLA

Divya D., et al. "Simulated Workbench Design for Characterisation and Selection of Appropriate Outlier Ensemble Algorithm." IJISMD vol.13, no.1 2022: pp.1-23. http://doi.org/10.4018/IJISMD.315024

APA

Divya D., Bhasi, M., & Kumar, S. (2022). Simulated Workbench Design for Characterisation and Selection of Appropriate Outlier Ensemble Algorithm. International Journal of Information System Modeling and Design (IJISMD), 13(1), 1-23. http://doi.org/10.4018/IJISMD.315024

Chicago

Divya D., M. Bhasi, and Santosh Kumar. "Simulated Workbench Design for Characterisation and Selection of Appropriate Outlier Ensemble Algorithm," International Journal of Information System Modeling and Design (IJISMD) 13, no.1: 1-23. http://doi.org/10.4018/IJISMD.315024

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

Outlier ensembles can enhance the performance of outlier detection algorithms. However, identification of an appropriate outlier ensemble algorithm for the given application is a difficult task. In order to achieve this function, the performance features of outlier ensembles with gathered data types containing a certain set of outliers have been characterised and their ensemble performance analysed in this paper. The proposed methodology suggests comparing data and anomalous features of a given real time application to the data set available in the proposed workbench. Users can use the results of experiments conducted using the workbench to identify highly efficient algorithms for their application without undertaking further analysis once matching data is located in the workbench. Therefore, this research paper act as a recommendation system that enables data scientists from various fields to pick an effective outlier ensemble for their application.