Melola Discovery

← Melola Discovery

Open access Jurnal SINTA 2

OPTIMIZING MACHINE LEARNING PIPELINE DESIGN THROUGH PROGRAMMING PARADIGM SELECTION

IJCCS (Indonesian Journal of Computing and Cybernetics Systems) · Vol. 20 No. 2 (2026) · pp. 163–174

DOI
10.22146/ijccs.112691
ISSN
2460-7258, 1978-1520
Published
2026-04-30
Publisher
IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.
Language
eng
License
creativecommons.org/licenses/by-sa/4.0

Abstract

This study investigates the impact of programming paradigm selection on the efficiency and sustainability of machine learning (ML) pipeline design. A case study was conducted using an agricultural IoT dataset for crop yield prediction, where four paradigms imperative, functional, object-oriented (OOP), and declarative were implemented to construct modular, maintainable, and reproducible pipelines. Each paradigm was evaluated through five key metrics: development time, debugging time, modularity, reproducibility, and maintainability. Experimental data were analyzed using descriptive statistics and visualized with boxplots and radar charts to identify performance differences. The results demonstrate that the functional paradigm achieved superior performance in data preprocessing with high reproducibility (95%), OOP produced the highest modularity (5.0/5), while the declarative paradigm exhibited the best reproducibility (98%) and deployment efficiency. In contrast, the imperative paradigm enabled faster prototyping but lacked long-term stability. Integrating paradigms in a multi-paradigm design reduced development time by 30.3%, debugging effort by 41.2%, and improved modularity and reproducibility by 41.6% and 21%, respectively. These findings highlight that no single paradigm is universally optimal; instead, a multi-paradigm approach provides a more efficient, maintainable, and production-ready ML pipeline framework adaptable to industrial-scale implementations.

Keywords

Collaboration network of the authors

6 authors · 6 co-authorship links

Hover or tap a name to highlight its co-authors. Double-click to open the author's page.

  • Universitas Teknologi Digital Indonesia
  • Affiliation unknown
  • Different organisations
  • Same organisation
  • Affiliation unknown
  • Thicker line = more joint articles · bigger circle = more articles
Show as table
Co-authorOrganisationJoint articles
Pujianto PujiantoUniversitas Teknologi Digital Indonesia1
Saryanto SaryantoAffiliation unknown1
Suryanto SuryantoUniversitas Teknologi Digital Indonesia1

Affiliations come from OpenAlex; journal OAI-PMH metadata carries none. Authors without an OpenAlex ID are matched by name, which can merge or split people.

Metadata source: OAI-PMH (metadata jurnal) https://journal.ugm.ac.id/ijccs/oai, first harvested 26 Okt 2026, last seen 26 Okt 2026.
Original metadata (oai_dc) 
<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/  http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
	<dc:title xml:lang="en-US">OPTIMIZING MACHINE LEARNING PIPELINE DESIGN THROUGH PROGRAMMING PARADIGM SELECTION</dc:title>
	<dc:creator>Kusjani, Adi</dc:creator>
	<dc:creator>Andriyani, Widyastuti</dc:creator>
	<dc:creator>Kristomo, Domy</dc:creator>
	<dc:subject xml:lang="en-US">Computer Science; Programming Paradigms</dc:subject>
	<dc:subject xml:lang="en-US">Machine Learning Pipeline; Programming Paradigm; Multi-Paradigm Design; Reproducibility; Software Modularity</dc:subject>
	<dc:description xml:lang="en-US">This study investigates the impact of programming paradigm selection on the efficiency and sustainability of machine learning (ML) pipeline design. A case study was conducted using an agricultural IoT dataset for crop yield prediction, where four paradigms imperative, functional, object-oriented (OOP), and declarative were implemented to construct modular, maintainable, and reproducible pipelines. Each paradigm was evaluated through five key metrics: development time, debugging time, modularity, reproducibility, and maintainability. Experimental data were analyzed using descriptive statistics and visualized with boxplots and radar charts to identify performance differences. The results demonstrate that the functional paradigm achieved superior performance in data preprocessing with high reproducibility (95%), OOP produced the highest modularity (5.0/5), while the declarative paradigm exhibited the best reproducibility (98%) and deployment efficiency. In contrast, the imperative paradigm enabled faster prototyping but lacked long-term stability. Integrating paradigms in a multi-paradigm design reduced development time by 30.3%, debugging effort by 41.2%, and improved modularity and reproducibility by 41.6% and 21%, respectively. These findings highlight that no single paradigm is universally optimal; instead, a multi-paradigm approach provides a more efficient, maintainable, and production-ready ML pipeline framework adaptable to industrial-scale implementations.</dc:description>
	<dc:publisher xml:lang="en-US">IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.</dc:publisher>
	<dc:contributor xml:lang="en-US">Adi Kusjani, Universitas Teknologi Digital Indonesia</dc:contributor>
	<dc:contributor xml:lang="en-US">Widyastuti Andriyani, Universitas Teknologi Digital Indonesia</dc:contributor>
	<dc:contributor xml:lang="en-US">Domy Kristomo, Universitas Teknologi Digital Indonesia</dc:contributor>
	<dc:contributor xml:lang="en-US"></dc:contributor>
	<dc:date>2026-04-30</dc:date>
	<dc:type>info:eu-repo/semantics/article</dc:type>
	<dc:type>info:eu-repo/semantics/publishedVersion</dc:type>
	<dc:type xml:lang="en-US"></dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:identifier>https://journal.ugm.ac.id/ijccs/article/view/112691</dc:identifier>
	<dc:identifier>10.22146/ijccs.112691</dc:identifier>
	<dc:source xml:lang="en-US">IJCCS (Indonesian Journal of Computing and Cybernetics Systems); Vol 20, No 2 (2026): April; 163-174</dc:source>
	<dc:source>2460-7258</dc:source>
	<dc:source>1978-1520</dc:source>
	<dc:language>eng</dc:language>
	<dc:relation>https://journal.ugm.ac.id/ijccs/article/view/112691/43999</dc:relation>
	<dc:rights xml:lang="en-US">Copyright (c) 2026 IJCCS (Indonesian Journal of Computing and Cybernetics Systems)</dc:rights>
	<dc:rights xml:lang="en-US">http://creativecommons.org/licenses/by-sa/4.0</dc:rights>
</oai_dc:dc>