eus.ixa.ixa.pipe.ml.utils

Class AnnotateProperties

java.lang.Object

eus.ixa.ixa.pipe.ml.utils.AnnotateProperties

public class AnnotateProperties
extends Object

Properties for annotation. Language and model parameters are compulsory.

1. Language: ca, de, en, es, eu, fr, gl, it, nl, pt, ru.
2. Model: the model name.

Version:

2016-04-22

Author:

ragerri

Field Summary

Fields

Modifier and Type	Field and Description
static String	DEFAULT_ALL_MORPHOLOGY Output all POS and Lemma analysis before disambiguation.
static String	DEFAULT_DICTAG Correct statistical POS tagger output with monosemic dictionary.
static String	DEFAULT_HARD_PARAGRAPH Choose between 'yes' and 'no'.
static String	DEFAULT_MULTIWORDS Detect multiwords.
static String	DEFAULT_NORMALIZE Choose corpus conventions for normalization of punctuation.
static String	DEFAULT_UNTOKENIZABLE_STRING Choose between 'yes' and 'no'.

Method Summary

Modifier and Type	Method and Description
Properties	setChunkingProperties(String model, String language) Generate Properties object for chunking.
static Properties	setPOSLemmaProperties(String model, String lemmatizerModel, String language, String multiwords, String dictag, String allMorphology) Generate Properties object for POS tagging and Lemmatizing.
static Properties	setTokenizeProperties(String lang, String normalize, String untokenizable, String hardParagraph) Creates the Properties object required to construct a Sentence Segmenter and a Tokenizer.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

DEFAULT_NORMALIZE

public static final String DEFAULT_NORMALIZE

Choose corpus conventions for normalization of punctuation. See eus.ixa.ixa.pipe.ml.tok.Normalizer class.

See Also:

Constant Field Values

DEFAULT_UNTOKENIZABLE_STRING

public static final String DEFAULT_UNTOKENIZABLE_STRING

Choose between 'yes' and 'no'.

See Also:

Constant Field Values

DEFAULT_HARD_PARAGRAPH

public static final String DEFAULT_HARD_PARAGRAPH

Choose between 'yes' and 'no'.

See Also:

Constant Field Values

DEFAULT_MULTIWORDS

public static final String DEFAULT_MULTIWORDS

Detect multiwords.

See Also:

Constant Field Values

DEFAULT_DICTAG

public static final String DEFAULT_DICTAG

Correct statistical POS tagger output with monosemic dictionary.

See Also:

Constant Field Values

DEFAULT_ALL_MORPHOLOGY

public static final String DEFAULT_ALL_MORPHOLOGY

Output all POS and Lemma analysis before disambiguation.

See Also:

Constant Field Values

Method Detail

setTokenizeProperties

public static Properties setTokenizeProperties(String lang,
                                               String normalize,
                                               String untokenizable,
                                               String hardParagraph)

Creates the Properties object required to construct a Sentence Segmenter and a Tokenizer.

Parameters:

lang - it is required to provide a language code

normalize - the normalization option

untokenizable - print untokenizable tokens

hardParagraph - do not segment paragraph marks

Returns:

the properties object

setPOSLemmaProperties

public static Properties setPOSLemmaProperties(String model,
                                               String lemmatizerModel,
                                               String language,
                                               String multiwords,
                                               String dictag,
                                               String allMorphology)

Generate Properties object for POS tagging and Lemmatizing. Language, model and lemmatizerModel are compulsory.

Parameters:

model - the pos tagger model

lemmatizerModel - the lemmatizer model

language - the language

multiwords - whether multiwords are to be detected

dictag - whether tagging from a dictionary is activated

allMorphology - whether to disclose all pos tags and lemmas before disambiguation

Returns:

a properties object

setChunkingProperties

public Properties setChunkingProperties(String model,
                                        String language)

Generate Properties object for chunking. Both parameters are compulsory.

Parameters:

model - the model to perform the annotation

language - the language

Returns:

a properties object