An Approach to Experiment Reproducibility Through MLOps and Semantic Web Technologies
This article addresses the challenge of reproducing machine learning (ML) experiments by integrating processes based on MLOps and semantic technologies. The inherent complexity of experimentation in scientific research hinders reproducibility through conventional methods, which has led to the need t...
| Main Authors: | , , , |
|---|---|
| Format: | ARTÍCULO DE CONFERENCIA |
| Language: | es_ES |
| Published: |
IEEE
2024
|
| Subjects: | |
| Online Access: | http://dspace.ucuenca.edu.ec/handle/123456789/44127 https://www.scopus.com/inward/record.uri?eid=2-s2.0-85182277987&doi=10.1109%2fCLEI60451.2023.10346140&partnerID=40&md5=5ce72fe9f4acef05dba4a92df36bf0a7 |
| _version_ | 1793400620815745024 |
|---|---|
| author | Seaman Mora, Daniel Andres Saquicela Galarza, Victor Hugo Palacio Baus, Kenneth Samuel Peñafiel Mora, David Marcelo |
| author2 | Seaman Mora, Daniel Andres |
| author_facet | Seaman Mora, Daniel Andres Seaman Mora, Daniel Andres Saquicela Galarza, Victor Hugo Palacio Baus, Kenneth Samuel Peñafiel Mora, David Marcelo |
| author_sort | Seaman Mora, Daniel Andres |
| collection | DSpace |
| description | This article addresses the challenge of reproducing machine learning (ML) experiments by integrating processes based on MLOps and semantic technologies. The inherent complexity of experimentation in scientific research hinders reproducibility through conventional methods, which has led to the need to automate processes. In this work, a solution has been developed allowing the execution of ML experiments of other researchers and their reproducibility. The use of semantic technologies allows the complete description of the experiment, including the data and resources necessary for its execution. The approach proposed in this work contributes to the automation of the experimentation phases based on MLOps, demonstrating how it can be used to reproduce experiments and offer a solution to the complexity of experimentation in scientific research. The effectiveness of the solution proposed in this work is evaluated by means of a survey-based analysis carried out among researchers who currently use manual processes to perform machine learning experiments. The results indicate that manual processing is prone to errors and not scalable regarding the size and complexity of most experiments. Moreover, the solution proposed in this work, which combines MLOps-based processes and semantic technologies, has been well received by researchers and considered to significantly improve the efficiency, reproducibility, and scalability of machine learning experimentation. |
| format | ARTÍCULO DE CONFERENCIA |
| id | oai:dspace.ucuenca.edu.ec:123456789-44127 |
| institution | Universidad de Cuenca |
| language | es_ES |
| publishDate | 2024 |
| publisher | IEEE |
| record_format | dspace |
| spelling | oai:dspace.ucuenca.edu.ec:123456789-441272024-03-06T19:03:09Z An Approach to Experiment Reproducibility Through MLOps and Semantic Web Technologies Seaman Mora, Daniel Andres Saquicela Galarza, Victor Hugo Palacio Baus, Kenneth Samuel Peñafiel Mora, David Marcelo Seaman Mora, Daniel Andres Experiment Machine learning MLOps Semantic Web Reproducibility This article addresses the challenge of reproducing machine learning (ML) experiments by integrating processes based on MLOps and semantic technologies. The inherent complexity of experimentation in scientific research hinders reproducibility through conventional methods, which has led to the need to automate processes. In this work, a solution has been developed allowing the execution of ML experiments of other researchers and their reproducibility. The use of semantic technologies allows the complete description of the experiment, including the data and resources necessary for its execution. The approach proposed in this work contributes to the automation of the experimentation phases based on MLOps, demonstrating how it can be used to reproduce experiments and offer a solution to the complexity of experimentation in scientific research. The effectiveness of the solution proposed in this work is evaluated by means of a survey-based analysis carried out among researchers who currently use manual processes to perform machine learning experiments. The results indicate that manual processing is prone to errors and not scalable regarding the size and complexity of most experiments. Moreover, the solution proposed in this work, which combines MLOps-based processes and semantic technologies, has been well received by researchers and considered to significantly improve the efficiency, reproducibility, and scalability of machine learning experimentation. La Paz 2024-03-06T19:03:06Z 2024-03-06T19:03:06Z 2023 ARTÍCULO DE CONFERENCIA 979-8-3503-1887-6 2771-5752 http://dspace.ucuenca.edu.ec/handle/123456789/44127 https://www.scopus.com/inward/record.uri?eid=2-s2.0-85182277987&doi=10.1109%2fCLEI60451.2023.10346140&partnerID=40&md5=5ce72fe9f4acef05dba4a92df36bf0a7 10.1109/CLEI60451.2023.10346140 es_ES application/pdf IEEE XLIX Latin American Computer Conference (CLEI) |
| spellingShingle | Experiment Machine learning MLOps Semantic Web Reproducibility Seaman Mora, Daniel Andres Saquicela Galarza, Victor Hugo Palacio Baus, Kenneth Samuel Peñafiel Mora, David Marcelo An Approach to Experiment Reproducibility Through MLOps and Semantic Web Technologies |
| title | An Approach to Experiment Reproducibility Through MLOps and Semantic Web Technologies |
| title_full | An Approach to Experiment Reproducibility Through MLOps and Semantic Web Technologies |
| title_fullStr | An Approach to Experiment Reproducibility Through MLOps and Semantic Web Technologies |
| title_full_unstemmed | An Approach to Experiment Reproducibility Through MLOps and Semantic Web Technologies |
| title_short | An Approach to Experiment Reproducibility Through MLOps and Semantic Web Technologies |
| title_sort | approach to experiment reproducibility through mlops and semantic web technologies |
| topic | Experiment Machine learning MLOps Semantic Web Reproducibility |
| url | http://dspace.ucuenca.edu.ec/handle/123456789/44127 https://www.scopus.com/inward/record.uri?eid=2-s2.0-85182277987&doi=10.1109%2fCLEI60451.2023.10346140&partnerID=40&md5=5ce72fe9f4acef05dba4a92df36bf0a7 |
| work_keys_str_mv | AT seamanmoradanielandres anapproachtoexperimentreproducibilitythroughmlopsandsemanticwebtechnologies AT saquicelagalarzavictorhugo anapproachtoexperimentreproducibilitythroughmlopsandsemanticwebtechnologies AT palaciobauskennethsamuel anapproachtoexperimentreproducibilitythroughmlopsandsemanticwebtechnologies AT penafielmoradavidmarcelo anapproachtoexperimentreproducibilitythroughmlopsandsemanticwebtechnologies AT seamanmoradanielandres approachtoexperimentreproducibilitythroughmlopsandsemanticwebtechnologies AT saquicelagalarzavictorhugo approachtoexperimentreproducibilitythroughmlopsandsemanticwebtechnologies AT palaciobauskennethsamuel approachtoexperimentreproducibilitythroughmlopsandsemanticwebtechnologies AT penafielmoradavidmarcelo approachtoexperimentreproducibilitythroughmlopsandsemanticwebtechnologies |