Automated Machine Learning Systems for End-to-End Model Design in Data-Centric Applications
Keywords:
AutoML, Data-Centric AI, Pipeline Automation, Model Optimization, End-to-End SystemsAbstract
Automated Machine Learning (AutoML) systems have revolutionized how machine learning models are developed and deployed, particularly in data-centric environments where data quality, integration, and transformation govern performance. This paper examines the evolution of AutoML frameworks and their effectiveness in supporting end-to-end workflows, from data preprocessing to model evaluation. Emphasis is placed on recent advances in pipeline optimization, data-centric AI, and minimal human-intervention strategies. We provide a synthesis of relevant literature, introduce a comparative analysis of leading systems, and propose a conceptual architecture for integrating AutoML into dynamic data environments. Our findings suggest that the fusion of data-centric principles with AutoML enhances model robustness, interpretability, and deployment efficiency.
References
Thornton, Chris, et al. "Auto-WEKA: Combined Selection and Hyperparameter Optimization of Classification Algorithms." Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2013, pp. 847–855.
Nagamani, N. (2024). AI-driven risk assessment models for health and life insurance underwriting. International Journal of Computer Science and Engineering (IACSE-IJCSE), 5(1). https://doi.org/10.5281/zenodo.17852768
Feurer, Matthias, et al. "Efficient and Robust Automated Machine Learning." Advances in Neural Information Processing Systems, vol. 28, 2015, pp. 2962–2970.
Olson, Randal S., et al. "Evaluation of a Tree-Based Pipeline Optimization Tool for Automating Data Science." Proceedings of the Genetic and Evolutionary Computation Conference, 2016, pp. 485–492.
Zoph, Barret, and Quoc V. Le. "Neural Architecture Search with Reinforcement Learning." International Conference on Learning Representations, 2017.
Hall, Patrick, et al. Machine Learning Pipelines: A New Understanding. O’Reilly Media, 2021.
Hutter, Frank, Lars Kotthoff, and Joaquin Vanschoren, editors. Automated Machine Learning: Methods, Systems, Challenges. Springer, 2019.
Nagamani, N. (2023). Hybrid AI Models Combining Financial NLP and Time-Series Forecasting for Stock Advisory. International Journal of Scientific Research in Artificial Intelligence and Machine Learning (ISCSITR-IJSRAIML), 4(1), 61–74. https://doi.org/10.63397/ISCSITR-IJSRAIML_2023_04_01_005
Elsken, Thomas, Jan Hendrik Metzen, and Frank Hutter. "Neural Architecture Search: A Survey." Journal of Machine Learning Research, vol. 20, no. 55, 2019, pp. 1–21.
He, Xin, et al. "AutoML: A Survey of the State-of-the-Art." Knowledge-Based Systems, vol. 212, 2021, 106622.
Nagamani, N. (2023). Predictive AI models for reducing payment failures in digital wallet systems. International Journal of Fintech (IJFT), 2(1), 7–20. https://doi.org/10.34218/IJFT_02_01_002
Yao, Quanming, et al. "Taking Human out of Learning Applications: A Survey on Automated Machine Learning." arXiv preprint, arXiv:1810.13306, 2018.
Truong, Anh, et al. "Towards Automated Machine Learning: Evaluation and Comparison of AutoML Approaches and Tools." Information, vol. 11, no. 4, 2020, p. 193.
Nagamani, N. (2024). Multi-layer AI defense models against real-time phishing and deepfake financial fraud. ISCSITR - International Journal of Business Intelligence (ISCSITR-IJBI), 5(2), 7–21. https://doi.org/10.63397/ISCSITR-IJBI_05_02_02.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Siddharth Nair K (Author)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
