Number of the records: 1
Using Artificial Neural Networks to Determine Ontologies Most Relevant to Scientific Texts
- 1.
SYSNO ASEP 0565961 Document Type C - Proceedings Paper (int. conf.) R&D Document Type Conference Paper Title Using Artificial Neural Networks to Determine Ontologies Most Relevant to Scientific Texts Author(s) Korel, L. (CZ)
Behr, A. S. (DE)
Kockmann, N. (DE)
Holeňa, Martin (UIVT-O) SAI, RIDNumber of authors 4 Source Title Proceedings of the 22st Conference Information Technologies – Applications and Theory (ITAT 2022). - Aachen : Technical University & CreateSpace Independent Publishing, 2022 / Ciencialová L. ; Holeňa M. ; Jajcay R. ; Jajcayová R. ; Mráz F. ; Pardubská D. ; Plátek M. - ISSN 1613-0073 Pages s. 44-54 Number of pages 11 s. Publication form Online - E Action ITAT 2022: Conference Information Technologies - Applications and Theory /22./ Event date 23.09.2022 - 27.09.2022 VEvent location Zuberec Country SK - Slovakia Event type EUR Language eng - English Country DE - Germany Keywords ontology ; text data ; text preprocessing ; text representation learning ; text classification OECD category Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8) Institutional support UIVT-O - RVO:67985807 EID SCOPUS 85139874007 Annotation This paper provides an insight into the possibility of how to find ontologies most relevant to scientific texts using artificial neural networks. The basic idea of the presented approach is to select a representative paragraph from a source text file, embed it to a vector space by a pre-trained fine-tuned transformer, and classify the embedded vector according to its relevance to a target ontology. We have considered different classifiers to categorize the output from the transformer, in particular random forest, support vector machine, multilayer perceptron, k-nearest neighbors, and Gaussian process classifiers. Their suitability has been evaluated in a use case with ontologies and scientific texts concerning catalysis research. From results we can say the worst results have random forest. The best results in this task brought support vector machine classifier Workplace Institute of Computer Science Contact Tereza Šírová, sirova@cs.cas.cz, Tel.: 266 053 800 Year of Publishing 2023 Electronic address https://ceur-ws.org/Vol-3226/paper5.pdf
Number of the records: 1