Navneet Shukla,
Aadil Siddiqui,
Padma Mishra,
- Research Scholar, Thakur Institute of Management Studies, Career Development & Research (TIMSCDR), Maharashtra, India
- Research Scholar, Thakur Institute of Management Studies, Career Development & Research (TIMSCDR), Maharashtra, India
- Associate Professor, Thakur Institute of Management Studies, Career Development & Research (TIMSCDR), Maharashtra, India
Abstract
Machine learning models such as Decision Trees, Logistic Regression, and K-Nearest Neighbors are widely used for classification tasks due to their simplicity and interpretability. However, these models often suffer from high variance, overfitting, and poor generalization when applied to real-world datasets, particularly those that are small, noisy, or imbalanced, as commonly encountered in healthcare, finance, and cybersecurity applications. To address these limitations, this research proposes a Bagging (Bootstrap Aggregating)-based ensemble learning approach aimed at improving the robustness, stability, and predictive performance of traditional machine learning models. The proposed methodology employs bootstrap sampling to generate multiple subsets of the training data and trains diverse base learners independently, with their predictions aggregated through majority voting. This ensemble strategy effectively reduces variance, minimizes overfitting, and enhances model generalization without significantly increasing computational complexity. The performance of the proposed approach is evaluated using three benchmark datasets—Diabetes, Iris, and Digits—and compared against individual machine learning models. Experimental results demonstrate that the Bagging-based ensemble consistently achieves higher classification accuracy, improved precision, enhanced recall, and lower prediction variance across all datasets. Furthermore, to improve accessibility and interpretability, a Streamlit-based interactive web application is developed, enabling users to visualize model performance, compare ensemble and standalone classifiers, and gain insights into the behavior of Bagging techniques. The findings highlight that Bagging serves as a lightweight, interpretable, and effective ensemble learning method capable of producing stable and reliable machine learning models. This work demonstrates its practical value for educational purposes as well as real-world predictive analytics, offering a simple yet powerful solution for improving classification performance across diverse application domains.
Keywords: Bagging, ensemble learning, model robustness, bootstrap aggregating, streamlit, machine learning stability
[This article belongs to Recent Trends in Mathematics ]
References
1. L. Breiman, “Bagging predictors,” Machine Learning, vol. 24, no. 2, pp. on standard classification benchmarks, many specialized domains 123–140, 1996.
2. T. Dietterich, “Ensemble methods in machine learning,” in Multiple Classifier Systems. Springer, 2000.
3. A. Liaw and M. Wiener, “Classification and regression by randomForest,” natural language processing, document-level or sentence-level boot- R News, vol. 2, no. 3, pp. 18–22, 2002.
4. T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” Proceedings of the ACM SIGKDD International Conference on KnowledgeDiscovery and Data Mining (KDD), 2016.
5. R. Polikar, “Ensemble based systems in decision making,” IEEE Circuits and Systems Magazine, vol. 6, no. 3, pp. 21–45, 2006.
6. L. Yang and A. Shami, “On hyperparameter optimization of machine learning algorithms,” arXiv preprint, arXiv:2007.15745, 2020.
7. R. Turner et al., “Bayesian optimization is superior to random search for hyperparameter tuning,” arXiv preprint, arXiv:2104.10201, 2021.
8. B. Efron, “Bootstrap methods: Another look at the jackknife,” The Annals of Statistics, vol. 7, no. 1, pp. 1–26, 1979.
9. J. Friedman, T. Hastie, and R. Tibshirani, The Elements of Statistical Learning. Springer Series in Statistics, 2001.
10. P. Gerela, P. N. Mishra, and R. Vipat, “Study on data visualization: Its importance in the education sector,” International Journal of Health Sciences, vol. 6, no. NS3, pp. 6298–6305, Apr. 2022, doi: 10.53730/ijhs.v6ns3.7393.
11. S. Nagarkar, P. Mishra, and V. Gaikwad, “Factors at play: Investigating the dimensions of privacy and security in smart home environments,” Educational Administration: Theory and Practice, vol. 30, no. 5, pp. 3197–3203, 2024, doi: 10.53555/kuey.v30i5.3414.
12. P. Mishra, V. Gaikwad, M. Paradkar, N. Pandey, R. Bansode, and S. Maitra, “Implication of analysis of machine learning models for predicting the risk of cardiovascular disease by considering lifestyle factors and feature selection,” Educational Administration: Theory and Practice, vol. 30, no. 5, pp. 8286–8298, 2024, doi: 10.53555/kuey.v30i5.4353.
13. P. Mishra, P. Gerala, and S. Maitra, “Study on artificial intelligence application uses in agriculture,” International Journal of Health Sciences, vol. 6, no. NS2, 2022, doi: 10.53730/ijhs.v6ns2.7391.
14. P. Mishra, S. Nagarkar, and V. Gaikwad, “Integrating variable rate application with machine learning algorithms for intelligent and responsive agriculture,” Journal of Technical Education, vol. 55, 2024.
15. G. N. Basavaraj, “Machine learning-enhanced direction-of-arrival estimation for coherent and non-coherent sources,” Engineering, Technology & Applied Science Research, vol. 15, no. 2, pp. 20647–20652, Apr. 2025.
16. E. Ahmad, B. Dash, A. Tripathi, et al., “Hybrid CNN and image processing framework for precise characterization of cracks in concrete structures,” Asian Journal of Civil Engineering, vol. 27, pp. 815–828, 2026, doi: 10.1007/s42107-025-01535-0.
17. V. Gaikwad, “AI-enabled early detection of fetal gestational age and CNS anomalies in the first trimester through ultrasound to support rural doctors in India,” SSRG International Journal of Electronics and Communication Engineering, vol. 12, no. 7, pp. 174–183, 2025, doi: 10.14445/23488549/IJECE-V12I7P113.
Recent Trends in Mathematics
| Volume | 03 | |
| Issue | 02 | |
| Received | 16/03/2026 | |
| Accepted | 24/07/2026 | |
| Published | 10/08/2026 | |
| Publication Time | 147 Days |