FAQ

Schedae Informaticae

logo of Jagiellonian University in Krakow

Evaluating KGR10 Polish Word Embeddings in the Recognition of Temporal Expressions Using BiLSTM-CRF

Publication date: 2018

Schedae Informaticae, 2018, Volume 27, pp. 93 - 106

https://doi.org/10.4467/20838476SI.18.008.10413

Authors

,
Jan Kocoń
Department of Computational Intelligence, Faculty of Computer Science and Management, Wroclaw University of Technology
All publications →
Michał Gawor
Department of Computational Intelligence, Faculty of Computer Science and Management, Wroclaw University of Technology
All publications →

Abstract

The article introduces a new set of Polish word embeddings, built using KGR10 corpus, which contains more than 4 billion words. These embeddings are evaluated in the problem of recognition of temporal expressions (timexes) for the Polish language. We described the process of KGR10 corpus creation and a new approach to the recognition problem using Bidirectional Long-Short Term Memory (BiLSTM) network with additional CRF layer, where specific embeddings are essential. We presented experiments and conclusions drawn from them.

References


Information

Information: Schedae Informaticae, 2018, pp. 93 - 106

Article type: Original research article

Titles:

English:

Evaluating KGR10 Polish Word Embeddings in the Recognition of Temporal Expressions Using BiLSTM-CRF

Authors

Department of Computational Intelligence, Faculty of Computer Science and Management, Wroclaw University of Technology

Department of Computational Intelligence, Faculty of Computer Science and Management, Wroclaw University of Technology

Published at: 2018

Article status: Open

Licence: CC BY-NC-ND  licence icon

Percentage share of authors:

Jan Kocoń (Author) - 50%
Michał Gawor (Author) - 50%

Article corrections:

-

Publication languages:

English

View count: 1761

Number of downloads: 0