MEX vocabulary: A lightweight interchange format for machine learning experiments
Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review
Standard
Proceedings of the 11th International Conference on Semantic Systems, SEMANTiCS 2015. ed. / Axel Polleres; Sebastian Hellmann; Josiane Xavier Parreira. Association for Computing Machinery, Inc, 2015. p. 169-176 (ACM International Conference Proceeding Series; Vol. 16-17-September-2015).
Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - CHAP
T1 - MEX vocabulary
T2 - 11th International Conference on Semantic Systems, SEMANTiCS 2015
AU - Esteves, Diego
AU - Moussallem, Diego
AU - Neto, Ciro Baron
AU - Soru, Tommaso
AU - Usbeck, Ricardo
AU - Ackermann, Markus
AU - Lehmann, Jens
N1 - Horizon 2020 Framework Programme Funding number: 644055
PY - 2015/9/16
Y1 - 2015/9/16
N2 - Over the last decades many machine learning experiments have been published, giving benefit to the scientific progress. In order to compare machine-learning experiment results with each other and collaborate positively, they need to be performed thoroughly on the same computing environment, using the same sample datasets and algorithm configurations. Besides this, practical experience shows that scientists and engineers tend to have large output data in their experiments, which is both difficult to analyze and archive properly without provenance metadata. However, the Linked Data community still misses a lightweight specification for interchanging machine-learning metadata over different architectures to achieve a higher level of interoperability. In this paper, we address this gap by presenting a novel vocabulary dubbed MEX. We show that MEX provides a prompt method to describe experiments with a special focus on data provenance and fulfills the requirements for a long-term maintenance.
AB - Over the last decades many machine learning experiments have been published, giving benefit to the scientific progress. In order to compare machine-learning experiment results with each other and collaborate positively, they need to be performed thoroughly on the same computing environment, using the same sample datasets and algorithm configurations. Besides this, practical experience shows that scientists and engineers tend to have large output data in their experiments, which is both difficult to analyze and archive properly without provenance metadata. However, the Linked Data community still misses a lightweight specification for interchanging machine-learning metadata over different architectures to achieve a higher level of interoperability. In this paper, we address this gap by presenting a novel vocabulary dubbed MEX. We show that MEX provides a prompt method to describe experiments with a special focus on data provenance and fulfills the requirements for a long-term maintenance.
KW - Data Provenance
KW - Interchange Format
KW - Machine Learning Experiments
KW - Vocabulary
KW - Informatics
KW - Business informatics
UR - http://www.scopus.com/inward/record.url?scp=84962485425&partnerID=8YFLogxK
U2 - 10.1145/2814864.2814883
DO - 10.1145/2814864.2814883
M3 - Article in conference proceedings
AN - SCOPUS:84962485425
T3 - ACM International Conference Proceeding Series
SP - 169
EP - 176
BT - Proceedings of the 11th International Conference on Semantic Systems, SEMANTiCS 2015
A2 - Polleres, Axel
A2 - Hellmann, Sebastian
A2 - Parreira, Josiane Xavier
PB - Association for Computing Machinery, Inc
Y2 - 16 September 2015 through 17 September 2015
ER -