Python wordnet.synset_from_pos_and_offset方法代码示例

# 需要导入模块: from nltk.corpus import wordnet [as 别名]
# 或者: from nltk.corpus.wordnet import synset_from_pos_and_offset [as 别名]
def fit(self, X):
        """Fit the transformer."""
        super().fit(X)
        X = self._unpack(X)

        related = Counter()
        for x in X:
            for (offset, pos) in x:
                s = wn.synset_from_pos_and_offset(pos, int(offset))
                related.update(self.recursive_related(s))

        if self.prune:
            related = {k: v for k, v in related.items() if v > 1}
        self.feature_names = set(chain.from_iterable(X))
        self.features = {k: idx for idx, k in enumerate(related)}
        self.vec_len = len(self.features)
        return self 

# 需要导入模块: from nltk.corpus import wordnet [as 别名]
# 或者: from nltk.corpus.wordnet import synset_from_pos_and_offset [as 别名]
def procecss_wn18_definitions(self, wn18_path=""):
        definitions_file = os.path.join(wn18_path, "wordnet-mlj12-definitions.txt")
        with open(definitions_file) as f:
            lines = f.readlines()

        pos_conversion_map = {'NN':'n', 'VB':'v', 'JJ':'a', 'RB':'r'}
        self.definition_map = {}
        n_empty_definitions = 0
        for line in tqdm(lines):
            synset_offset, synset_tag, _ = line.split('\t')
            #definition_map[synset_idx.strip()] = definition.strip()
            #fetch definition from wordnet
            pos = synset_tag.split('_')[-2]
            wn_ss = wn.synset_from_pos_and_offset(pos_conversion_map[pos], int(synset_offset))
            definition = wn_ss.definition().strip()
            if len(definition) == 0:
                n_empty_definitions = n_empty_definitions + 1

            self.definition_map[synset_offset.strip()] = definition

        print ("#Empty definitions {}/{}".format(n_empty_definitions, len(self.definition_map)))
        synsets = sorted(self.definition_map.keys())
        self.synset_to_idx = {v:i for i,v in enumerate(synsets)}
        self.idx_to_synset = {v:i for i,v in self.synset_to_idx.items()}
        self.definitions = [self.definition_map[k] for k in synsets] 

# 需要导入模块: from nltk.corpus import wordnet [as 别名]
# 或者: from nltk.corpus.wordnet import synset_from_pos_and_offset [as 别名]
def senti_synset(self, *vals):
        from nltk.corpus import wordnet as wn

        if tuple(vals) in self._db:
            pos_score, neg_score = self._db[tuple(vals)]
            pos, offset = vals
            if pos == 's':
                pos = 'a'
            synset = wn.synset_from_pos_and_offset(pos, offset)
            return SentiSynset(pos_score, neg_score, synset)
        else:
            synset = wn.synset(vals[0])
            pos = synset.pos()
            if pos == 's':
                pos = 'a'
            offset = synset.offset()
            if (pos, offset) in self._db:
                pos_score, neg_score = self._db[(pos, offset)]
                return SentiSynset(pos_score, neg_score, synset)
            else:
                return None 

# 需要导入模块: from nltk.corpus import wordnet [as 别名]
# 或者: from nltk.corpus.wordnet import synset_from_pos_and_offset [as 别名]
def id2ss(self,ID):
        """
        Parameters
            ----------
            ID : str
                the id of a chinese word found in zh_wordnet.
        Returns
            ----------
            nltk.corpus.reader.wordnet.Synset
                an object in en_wordnet.
        """
        return wn.synset_from_pos_and_offset(str(ID[-1:]), int(ID[:8])) 

# 需要导入模块: from nltk.corpus import wordnet [as 别名]
# 或者: from nltk.corpus.wordnet import synset_from_pos_and_offset [as 别名]
def vectorize(self, x):
        """Vectorize a word."""
        vec = np.zeros(self.vec_len)
        for (offset, pos) in x:
            s = wn.synset_from_pos_and_offset(pos, int(offset))
            res = self.recursive_related(s)
            if self.prune:
                res = [x for x in res if x in self.features]
            s = [self.features[s] for s in res]
            vec[s] = 1

        return vec 

# 需要导入模块: from nltk.corpus import wordnet [as 别名]
# 或者: from nltk.corpus.wordnet import synset_from_pos_and_offset [as 别名]
def getnode(x):
    return wn.synset_from_pos_and_offset('n', int(x[1:])) 

# 需要导入模块: from nltk.corpus import wordnet [as 别名]
# 或者: from nltk.corpus.wordnet import synset_from_pos_and_offset [as 别名]
def all_senti_synsets(self):
        from nltk.corpus import wordnet as wn

        for key, fields in self._db.items():
            pos, offset = key
            pos_score, neg_score = fields
            synset = wn.synset_from_pos_and_offset(pos, offset)
            yield SentiSynset(pos_score, neg_score, synset)