New methods for detecting anomalies in the Big Data were proposed
The article “Anomaly detection in big data based on clustering” "(DOI: 10.19139 / soic.v5i4.365), which is co-authored by the academician secretary of ANAS, Director of the Institute of Information Technology, academician Rasim Alguliev, head of the department, corresponding member of ANAS, Doctor of Technical Sciences Ramiz Aliguliyev and senior research fellow, PhD on Engineering Lyudmila Sukhostat was published in the Journal of “Statistics, Optimization & Information Computing’, published by the US Department of “International Academic Press”.
One of the major problems in detecting anomaly in Big data is to select a suitable method or to offer a new effective method. To this end, the article proposed hybrid clustering methods for detecting anomalies in Big data. Experiments on different dimensional bases were performed to evaluate the proposed methods and the results were compared with the results of the k-method method. 5 criteria were used in the article to evaluate the results of the methods. The results of the experiment showed that the proposed methods demonstrate higher accuracy compared to the k-method method. It should be noted here that the proposed methods can also be used for other purposes.
The article is based on a grant project called "Development of methods and algorithms for ensuring information security in Big data funded by the Science Development Fundation under the President of the Republic of Azerbaijan(Grant # EIF-KETPL-2-2015-1 25) - 56/05/1).
"Statistics, Optimization & Information Computing" is indexed and referenced in Scopus (Elsevier), Crossref, Google Scholar, DOAJ and other international scientific databases.