Mblem Class Reference

#include <mblem_mod.h>

Public Member Functions
	Mblem (TiCC::LogStream *, TiCC::LogStream *=0)
	create a Timbl based lemmatizer More...
	~Mblem ()
bool	init (const TiCC::Configuration &)
void	add_provenance (folia::Document &, folia::processor *) const
void	Classify (frog_record &)
void	Classify (const icu::UnicodeString &)
std::vector< std::pair< std::string, std::string > >	getResult () const
std::string	getTagset () const
std::string	version () const
void	filterTag (const std::string &)
void	makeUnique ()
void	add_lemmas (const std::vector< folia::Word * > &, const frog_data &) const

Constructor & Destructor Documentation

Mblem()

Mblem::Mblem ( TiCC::LogStream *  errlog, TiCC::LogStream *  dbglog = 0  )

explicit

create a Timbl based lemmatizer

Parameters

errlog	a LogStream for errors
dbglog	a LogStream for debugging

~Mblem()

Mblem::~Mblem

(

)

Member Function Documentation

add_lemmas()

void Mblem::add_lemmas	(	const std::vector< folia::Word * > &	wv,
		const frog_data &	fd
	)		const

add the lemma from 'fd' to the FoLiA list of Word

Parameters

wv	The folia:Word vector
fd	the folia_data with added lemmatizer results

add_provenance()

void Mblem::add_provenance	(	folia::Document &	doc,
		folia::processor *	main
	)		const

add provenance information to the FoLiA document

Parameters

doc	the foLiA document we are working on
main	the main processor (presumably Frog) we want to add a new one to

Classify() [1/2]

void Mblem::Classify

(

const icu::UnicodeString &

uWord

)

give the lemma for 1 word

Parameters

uWord

a Unicode string with the word the mblemResult struct will be filled with 1 or more (alternative) solutions of a lemma + a POS-tag

Classify() [2/2]

void Mblem::Classify

(

frog_record &

fd

)

add lemma information to the frog_data

Parameters

fd	The frog_data

this handles some special cases like ABBREVIATION, the token-strip rules and the one-one rules. All 'normal' cases are handled over to the Timbl classifier

filterTag()

void Mblem::filterTag

(

const std::string &

postag

)

filater all non-matching tags out of the mblem results

Parameters

postag

the tag, given by the CGN-tagger, that should match

Mblem produces a range of possible solutions with tags. We use the POS tag given by the CGN tagger to remove all solutions with a different tag

getResult()

vector< pair< string, string > > Mblem::getResult

(

)

const

extract the results into a list of lemma/tag pairs

getTagset()

std::string Mblem::getTagset ( ) const

inline

init()

bool Mblem::init

(

const TiCC::Configuration &

config

)

initialize the lemmatizer using the config

Parameters

config

the Configuration to use

Returns

true when no problems are detected

makeUnique()

void Mblem::makeUnique

(

)

filter out all results that are equal

version()

std::string Mblem::version ( ) const

inline

---

The documentation for this class was generated from the following files:

• include/frog/mblem_mod.h
• src/mblem_mod.cxx