Manuscript Number : SHISRRJ24727
Construction of a Meta-Learner for Unsupervised Anomaly Detection
Authors(2) :-S. E. Suresh, Chennuru Srihari Many real-world applications, such as network security and medical and health equipment, Unsupervised identification of anomalies. Given the wide range of unique situations associated with AD work, no single approach has been demonstrated to be superior to the others. Academics have been particularly interested in the Algorithm Selection Problem (ASP), also known as algorithm selection, when it comes to supervised classification issues employing AutoML and meta-learning; unsupervised AD tasks, on the other hand, have gotten less attention. This work presents a novel meta-learning technique that generates an efficient unsupervised AD algorithm given a set of meta-features extracted from the unlabeled input dataset. It is discovered that the recommended meta-learner outperforms the state-of-the-art option.
S. E. Suresh Model Selection, Unsupervised Identification of Anomalies, Meta-Learning, And Meta-Features. Publication Details Published in : Volume 7 | Issue 2 | March-April 2024 Article Preview
Assistant Professor, Department of MCA, Annamacharya Institute of Technology & Sciences, Tirupati, Andhra Pradesh, India
Chennuru Srihari
Post Graduate, Department of MCA, Annamacharya Institute of Technology & Sciences, Tirupati, Andhra Pradesh, India
Date of Publication : 2024-04-30
License: This work is licensed under a Creative Commons Attribution 4.0 International License.
Page(s) : 43-49
Manuscript Number : SHISRRJ24727
Publisher : Shauryam Research Institute
URL : https://shisrrj.com/SHISRRJ24727