Version: 0.0-6
Title: R/KEA Interface
Description: An R interface to KEA (Version 5.0). KEA (for Keyphrase Extraction Algorithm) allows for extracting keyphrases from text documents. It can be either used for free indexing or for indexing with a controlled vocabulary. For more information see http://www.nzdl.org/Kea/.
Imports: RKEAjars (≥ 5.0-1), rJava (≥ 0.6-3), tm
SystemRequirements: Java (>= 5.0)
License: GPL-2
Packaged: 2015-04-03 15:15:00 UTC; hornik
Author: Ingo Feinerer [aut], Kurt Hornik [aut, cre]
Maintainer: Kurt Hornik <Kurt.Hornik@R-project.org>
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2015-04-03 17:27:56

Create a KEA Model

Description

Create a keyphrase extraction model.

Usage

createModel(corpus, keywords, model, voc = "none", vocformat = "")

Arguments

corpus

A list of character vectors containing the text documents, e.g., a Corpus object as provided by package tm.

keywords

A list of character vectors containing the keywords for each document in corpus.

model

A character giving the path where the created model should be stored.

voc

A character pointing to a controlled vocabulary.

vocformat

A character giving the format of voc.

Details

A tutorial on keyword extraction is located at http://www.nzdl.org/Kea/Download/Kea-5.0-Readme.txt. There you can find details on the internals of KEA, including various parameter settings (e.g., details on vocabularies and supported formats for these).

When controlled vocabularies are used (by default: no), the voc argument should give the file path to the respective files without their extensions. When vocformat is "skos", the extension must be ‘.rdf’; when "text", there must be files with extensions ‘.en’, ‘.rel’ and ‘.use’.

Value

Invisibly returns model, i.e., the path to the created KEA model.

Author(s)

Ingo Feinerer

References

http://www.nzdl.org/Kea/

See Also

extractKeywords

Extract Keywords

Description

Extract keywords from text documents.

Usage

extractKeywords(corpus, model, voc = "none", vocformat = "")

Arguments

corpus

A list of character vectors containing the text documents, e.g., a Corpus object as provided by package~tm, used for keyword extraction.

model

A character giving the path to a KEA model.

voc

A character pointing to a controlled vocabulary.

vocformat

A character giving the format of voc.

Details

A tutorial on keyword extraction is located at http://www.nzdl.org/Kea/Download/Kea-5.0-Readme.txt. There you can find details on the internals of KEA, including various parameter settings (e.g., valid arguments for voc and vocformat).

Value

A list of character vectors corresponding to the keywords in corpus.

Author(s)

Ingo Feinerer

References

http://www.nzdl.org/Kea/

See Also

createModel