Romanian Journal of Information Science and Technology (ROMJIST)

An open – access publication

  |  HOME  |   GENERAL INFORMATION  |   ROMJIST ON-LINE  |  KEY INFORMATION FOR AUTHORS  |   COMMITTEES  |  

ROMJIST is a publication of Romanian Academy,
Section for Information Science and Technology

Editor – in – Chief:
Radu-Emil Precup

Honorary Co-Editors-in-Chief:
Horia-Nicolai Teodorescu
Gheorghe Stefan

Secretariate (office):
Adriana Apostol
Adress for correspondence: romjist@nano-link.net (after 1st of January, 2019)

Founding Editor-in-Chief
(until 10th of February, 2021):
Dan Dascalu

Editing of the printed version: Mihaela Marian (Publishing House of the Romanian Academy, Bucharest)

Technical editor
of the on-line version:
Lucian Milea (University POLITEHNICA of Bucharest)

Sponsor:
• National Institute for R & D
in Microtechnologies
(IMT Bucharest), www.imt.ro

ROMJIST Volume 21, No. 4, 2018, pp. 446-459
 

Aliaksei KOLESAU, Dmitrij ŠEŠOK, Mindaugas RYBOKAS
A Character-Based Part-of-Speech Tagger with Feedforward Neural Networks

ABSTRACT: This article presents a simple method to perform part-of-speech (POS) tagging with feedforward neural networks applied to learnable character embeddings. The motivation of the research is based on the fact that for some languages a human can find out the part of speech for a word just by its spelling even without knowing the meaning of the word (see C. Fries’s example “woggles ugged diggles”). One of the goals was to achieve high accuracy tagging without using semantic information (e.g. without word embeddings). This allows performing tagging for out of vocabulary words. Also, the dependency of the performance from context size was studied. The plausibility of the method was proved by building a POS-tagger with the accuracy comparable to state-of-the-art results

KEYWORDS: POS tagging, character embeddings, feedforward neural networks

Read full text (pdf)






  |  HOME  |   GENERAL INFORMATION  |   ROMJIST ON-LINE  |  KEY INFORMATION FOR AUTHORS  |   COMMITTEES  |