Latvian morphological tagger

50 Last view: 2020-10-07

Latvian morphological tagger

The Latvian tagger analyzes sentences in text and assigns to each token string of morphosyntactic tags and base form. Output is provided in XCES, TreeTagger or Moses compatible form. The tagset used is derived from MULTEXT-East format and consists of 28 categories.

You don’t have the permission to edit this resource.

DistributionAvailability

Available - Restricted Use

Contact Person

Aivars Bērziņš

toolService

Tool

Language Dependent

Input

Media type: Text

Resource type: Corpus

Modality: Written Language

Language: Latvian

Character encoding: UTF - 8

Output

Media type: Text

Resource type: Corpus

Modality: Written Language

Language: Latvian

Character encoding: UTF - 8

Annotation type: Lemmatization, Morphosyntactic Annotation - Pos Tagging

Annotation format: Multext-East derived

Segmentation level: Word

Conformance to standards best practices: MULTEXT, XCES

Operation

Operating system: Windows

Evaluation

Evaluated: True

Evaluation level: Technological

Evaluation type: Black Box

Evaluation measure: Automatic

Metadata

Created: 08/01/2013

Last Updated: 08/01/2013

Version

Version: v1.0

Usage

Actual Use - Nlp Applications

Use NLP Specific: Machine Translation, Morphosyntactic Tagging, Named Entity Recognition, Terminology Extraction, Web Services

Documentation

Document Type: Article

Mārcis Pinnis, Kārlis Goba, Maximum Entropy Model for Disambiguation of Rich Morphological Tags., 100 , pp. 14-22 , Systems and Frameworks for Computational Morphology - Second International Workshop, SFCM 2011 , 2011

Publisher: Springer

Book Title: SFCM

ISBN: 978-3-642-23137-7

Keywords: tagger, maximum entropy, inflective languages, Estonian, Latvian, Lithuanian

Document Language: English

People who looked at this resource also viewed the following: