Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM64119
Volume:
10 |
5 |
May 2026
Pages:
1-3
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Master of computer application
Modern college of engineering, Pune
Master of computer application
Modern college of engineering, Pune
While data preprocessing typically accounts for 50% to 80% of the effort in data science projects, it remains a largely ungoverned phase of the project lifecycle. Modern Automated Machine Learning (AutoML) tools prioritize predictive accuracy but often fail to provide the transparency and accountability required in regulated industries, leading to a "black-box" challenge. This study introduces a decision-centric five-layer framework designed to bridge this gap by combining YAML-based rule governance with granular decision provenance. By shifting from informal ad-hoc scripts to a formalized system of governed rules, the proposed architecture ensures that every data transformation is reproducible, justifiable and auditable. Experimental results across healthcare, finance, and climate science datasets indicate that the framework achieves total reproducibility and meets stringent regulatory requirements with minimal impact on computational performance.
KEYWORDS
The Governed Rule-Based Preprocessing Framework mitigates the risks associated with "black-box" data preparation by embedding decision provenance directly into the analytical pipeline. By standardizing and formalizing rule-based governance, this approach allows researchers and industry practitioners to perform data preparation that is both computationally efficient and legally defensible. Future research will explore the application of this framework to multimodal and real-time streaming data.
[1] A. Chapman et al., "Capturing and querying fine-grained provenance of preprocessing pipelines," VLDB, 2020. [2] A. Chapman et al., "Supporting Better Insights of Data Science Pipelines with Fine-grained Provenance," ACM TODS, 2024. [3] R. A. Sola Martínez et al., "Data preprocessing workflow for exhaled breath analysis," Sci. Rep., 2020. [4] S. Krishnan and E. Wu, "AlphaClean: Automatic Generation of Data Cleaning Pipelines," arXiv, 2019. [5] N. Parulian and B. Ludaescher, "Trust the Process: Analyzing Prospective Provenance for Data Cleaning," WWW, 2023. [6] A. Wagner, "FAIRly big: A framework for computationally reproducible processing," Sci. Data, 2022. [7] A. R. Hevner et al., "Design Science in Information Systems Research," MIS Quarterly, 2004. [8] S. Siddiqi et al., "SAGA: A Scalable Framework for Optimizing Data Cleaning Pipelines," 2023. [9] A. K. Ojha, "Reproducibility Under Preprocessing Variability," Zenodo, 2025. [10] "Health information systems: W3C PROV-DM instantiation," Book Chapter, 2022.
Publication Date
5/27/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
3
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...