A Comparative Study of Stacked Ensemble and SVM Models in Breast Cancer Classification

View at publisher Summarise with VIRAT
Authors: S. Nithya Selvakumari
Cited by: 0
References: 15

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The research evaluates Support Vector Machine (SVM) alongside a Stacked Ensemble Model for their diagnostic use in breast cancer through direct comparison. The two models learned from classification data consisting of malignant and benign tumor patterns to determine their performance capabilities. An SVM model achieved successful detection of malignant cases with 90.69 % accuracy alongside precision at 1.00 and recall of 78.95 % combined with an F1-score value of 88.23 %. Layered / stacking an ensemble learning algorithm that uses meta-model to stack or layer-wise stack many classification, or regression models together. The level- 0 predictions given in base models are fed into a more comprehensive model (level-1 or meta-model) which produces the end prediction produced the best results which delivered 94.18 % accuracy along with 0.96 precision and 91.67 % recall and 93.75% F1-score. The experimental findings from these results demonstrate ensemble learning delivers improved performance through stable operation while using generalized algorithms than independent SVM classification models. These acquired data points validate that ensemble methods enhance diagnostic support systems for breast cancer detection through optimal optimization features. The enhanced performance of the stacked ensemble model directly supports global efforts to combat life-threatening diseases by providing reliable, technology-driven healthcare solutions.

Fields of Study & Subjects

Artificial Intelligence in Healthcare AI in cancer detection Imbalanced Data Classification Techniques

Publication details

Journal / source
International Conference on Communication, Computing and Electronics Systems (ICCCES)
Pages
1284–1291
Publication date
21 January 2026
Language
EN