Work

An adaptive ensemble feature selection technique for model-agnostic diabetes prediction

Year: 2025

Type: article

Abstract: Ensemble learning aggregates several models' outputs to improve the overall model's performance. Ensemble feature selection separating the appropriate features from the extra and non-essential features. In this paper, the main focus will be to expand the scope of Ensemble Learning to include Feature Selection. We will propose an Ensemble Feature Selection Method called AdaptDiabfor Diabetes Prediction that is Model-Agnostic. Our approach combines diverse feature selection techniques, such as filter and wrapper methods, harnessing their complementary strengths. We have used an adaptive combiner function, which dynamically selects the most informative features based on the characteristics of the ensemble members. We demonstrate the effectiveness of our proposed AdaptDiab method through empirical studies using various classification models. Empirical Results of Our Proposed Ensemble Feature Selection Model outperforms traditional methods. This paper contributes to Ensemble Learning Methods and provides a Practical and Better Framework for Feature selection.

Source: Scientific Reports

DOI: 10.1038/s41598-025-91282-8

Authors: K. Natarajan, Dhanalakshmi Baskaran, Selvakumar Kamalanathan

Institutions: National Institute of Technology Tiruchirappalli, B.S. Abdur Rahman Crescent Institute of Science & Technology

Language: English

Topic: Artificial Intelligence in Healthcare

Subfield: Health Information Management

Field: Health Professions

Domain: Health Sciences

SDG: Good health and well-being

Open Access status: gold

Estimated APC paid: $2,690