Source code for nltk.corpus.reader.sentiwordnet

# -*- coding: utf-8 -*-
# Natural Language Toolkit: SentiWordNet
# Copyright (C) 2001-2019 NLTK Project
# Author: Christopher Potts <>
# URL: <>
# For license information, see LICENSE.TXT

An NLTK interface for SentiWordNet

SentiWordNet is a lexical resource for opinion mining.
SentiWordNet assigns to each synset of WordNet three
sentiment scores: positivity, negativity, and objectivity.

For details about SentiWordNet see:

    >>> from nltk.corpus import sentiwordnet as swn
    >>> print(swn.senti_synset('breakdown.n.03'))
    <breakdown.n.03: PosScore=0.0 NegScore=0.25>
    >>> list(swn.senti_synsets('slow'))
    [SentiSynset('decelerate.v.01'), SentiSynset('slow.v.02'),
    SentiSynset('slow.v.03'), SentiSynset('slow.a.01'),
    SentiSynset('slow.a.02'), SentiSynset('dense.s.04'),
    SentiSynset('slow.a.04'), SentiSynset('boring.s.01'),
    SentiSynset('dull.s.08'), SentiSynset('slowly.r.01'),
    >>> happy = swn.senti_synsets('happy', 'a')
    >>> happy0 = list(happy)[0]
    >>> happy0.pos_score()
    >>> happy0.neg_score()
    >>> happy0.obj_score()

import re
from nltk.compat import python_2_unicode_compatible
from nltk.corpus.reader import CorpusReader

[docs]@python_2_unicode_compatible class SentiWordNetCorpusReader(CorpusReader): def __init__(self, root, fileids, encoding='utf-8'): """ Construct a new SentiWordNet Corpus Reader, using data from the specified file. """ super(SentiWordNetCorpusReader, self).__init__(root, fileids, encoding=encoding) if len(self._fileids) != 1: raise ValueError('Exactly one file must be specified') self._db = {} self._parse_src_file() def _parse_src_file(self): lines =[0]).read().splitlines() lines = filter((lambda x: not"^\s*#", x)), lines) for i, line in enumerate(lines): fields = [field.strip() for field in re.split(r"\t+", line)] try: pos, offset, pos_score, neg_score, synset_terms, gloss = fields except: raise ValueError('Line %s formatted incorrectly: %s\n' % (i, line)) if pos and offset: offset = int(offset) self._db[(pos, offset)] = (float(pos_score), float(neg_score))
[docs] 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
[docs] def senti_synsets(self, string, pos=None): from nltk.corpus import wordnet as wn sentis = [] synset_list = wn.synsets(string, pos) for synset in synset_list: sentis.append(self.senti_synset( sentis = filter(lambda x: x, sentis) return sentis
[docs] 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)
[docs]@python_2_unicode_compatible class SentiSynset(object): def __init__(self, pos_score, neg_score, synset): self._pos_score = pos_score self._neg_score = neg_score self._obj_score = 1.0 - (self._pos_score + self._neg_score) self.synset = synset
[docs] def pos_score(self): return self._pos_score
[docs] def neg_score(self): return self._neg_score
[docs] def obj_score(self): return self._obj_score
def __str__(self): """Prints just the Pos/Neg scores for now.""" s = "<" s += + ": " s += "PosScore=%s " % self._pos_score s += "NegScore=%s" % self._neg_score s += ">" return s def __repr__(self): return "Senti" + repr(self.synset)