Acta Marisiensis.
Seria Technologica



ISSN 2668-4217
ISSN-L 2668-4217
(Online)


Română

HomeEditorial boardSubmit paperPublication ethicsContactIndexing
Year 2024
Volume 21 (XXXVIII), no 1

Year 2023
Volume 20 (XXXVII), no 1
Volume 20 (XXXVII), no 2

Year 2022
Volume 19 (XXXVI), no 1
Volume 19 (XXXVI), no 2

Year 2021
Volume 18 (XXXV), no 1
Volume 18 (XXXV), no 2

Year 2020
Volume 17 (XXXIV), no 1
Volume 17 (XXXIV), no 2

Year 2019
Volume 16 (XXXIII), no 1
Volume 16 (XXXIII), no 2

Year 2018
Volume 15 (XXXII), no 1
Volume 15 (XXXII), no 2

Year 2017
Volume 14 (XXXI), no 1
Volume 14 (XXXI), no 2

Year 2016
Volume 13 (XXX), no 1
Volume 13 (XXX), no 2

Year 2015
Volume 12 (XXIX), no 1
Volume 12 (XXIX), no 2

Year 2014
Volume 11 (XXVIII), no 1
Volume 11 (XXVIII), no 2

Year 2013
Volume 10 (XXVII), no 1
Volume 10 (XXVII), no 2

Year 2012
Volume 9 (XXVI), no 1
Volume 9 (XXVI), no 2

Year 2011
Volume 8 (XXV), no 1
Volume 8 (XXV), no 2

Year 2010
Volume 7 (XXIV), no 1
Volume 7 (XXIV), no 2

Year 2009
Volume 6 (XXIII)

2021, Volume 18 (XXXV), no 1

Efficiency Analysis of Deeplearning4J Neural Network Classifiers in Development of Transition Based Dependency Parsers

Author(s):
László CSÉPÁNYI-FÜRJES, László KOVÁCS, University of Miskolc Institute of Information Science, Miskolc-Egyetemváros, Hungary

Abstract:
Dependency parsing is a complex process in natural language text processing, text to semantic transformation. The efficiency improvement of dependency parsing is a current and an active research area in the NLP community. The paper presents four transitionbased dependency parser models with implementation using DL4J classifiers. The efficiency of the proposed models were tested with Hungarian language corpora. The parsing model uses a data representation form based on lightweight embedding and a novel morphological-description-vector format is proposed for the input layer. Based on the test experiments on parsing Hungarian text documents, the proposed list-based transitions parsers outperform the widespread stack-based variants.

DOI: https://doi.org/10.2478/amset-2021-0006

Pages: 33-39

View full article
Update: 19-Jun-2024 © Published by University Press