camel_tools.disambig.common

This sub-module contains common functions and classes used for disambiguation.

Classes

class camel_tools.disambig.common.ScoredAnalysis(score, analysis, diac, pos_lex_logprob, lex_logprob)

A named tuple containing an analysis and its score.

score

The overall score of the analysis.

analysis

The analysis dictionary. See CAMeL Morphology Features for more information on features and their values.

diac

The diactrized form of the associated analysis. Used for tie-breaking equally scored analyses.

pos_lex_log_prob

The log (base 10) of the probability of the associated pos-lex pair values. Used for tie-breaking equally scored analyses.

lex_log_prob

The log (base 10) of the probability of the associated lex value. Used for tie-breaking equally scored analyses.

class camel_tools.disambig.common.DisambiguatedWord(word, analyses)

A named tuple containing a word and a sorted list (from high to low score) of scored analyses.

word

The word being disambiguated.

analyses

List of scored analyses sorted from highest to lowest disambiguation score.

class camel_tools.disambig.common.Disambiguator

Abstract base class that all disambiguators should implement.

abstract all_feats()

Return a set of all features produced by this disambiguator.

abstract disambiguate(sentence, top=1)

Disambiguate words in a sentence.

Parameters:

sentence (list of str) – list of words representing a sentence to be disambiguated.
top (int, optional) – The number of top analyses to return. If set to zero or less, then all analyses are returned. Defaults to 1.

Returns: list of DisambiguatedWord: List of disambiguted words in sentence.

abstract disambiguate_word(sentence, word_ndx, top=1)

Disambiguate a word at a given index in a sentence.

Parameters:

sentence (list of str) – list of words representing a sentence.
word_ndx (int) – the index of the word in sentence to disambiguate.
top (int, optional) – The number of top analyses to return. If set to zero or less, then all analyses are returned. Defaults to 1.

Returns: DisambiguatedWord: The disambiguated word at index word_ndx in sentence.

abstract tok_feats()

Return a set of tokenization features produced by this disambiguator.