A hybrid wrapper approach for optimal feature selection based on a novel multiobjective technique
by I. Preethi; K. Dharmarajan
International Journal of System of Systems Engineering (IJSSE), Vol. 13, No. 3, 2023

Abstract: Recent technological improvements have caused data management challenges. In machine learning and data mining, preprocessing removes noisy, irrelevant, and redundant input. Feature selection is a pre-processing method that increases classifier performance. Researchers have employed single-objective optimisation to improve categorisation. Single-objective optimisation reduces accuracy. In this study, a hybrid wrapper approach feature selection employing recursive feature elimination (RFE) and extreme learning classifier (HWFS-RTE) algorithm is used to anticipate sickness. A multiobjective technique of improved teaching learning-based optimisation (iTLBO) is combined with the RFE method choose categorisation features. iTLBO also performs disease classification with the extreme learning machine (ELM) by tuning ELM's parameters using iTLBO. TLBO is a population-based optimisation strategy, as it requires more rounds to obtain the classification's fitness value. It combines extreme machine learning (ELM) to improve convergence speed and reduce computing complexity. In disease prediction, the hybrid model (HWFS-RTE) applied to the Breast cancer dataset from the UCI repository has a 98.4% accuracy rate.

Online publication date: Thu, 24-Aug-2023

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 System of Systems Engineering (IJSSE):
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