Python Gensim word2vec词汇密钥

Question

I want to make word2vec with gensim. I heard that vocabulary corpus should be unicode so I converted it to unicode.

# -*- encoding:utf-8 -*-
# !/usr/bin/env python
import sys
reload(sys)
sys.setdefaultencoding('utf-8')
from gensim.models import Word2Vec
import pprint

with open('parsed_data.txt', 'r') as f:
    corpus = map(unicode, f.read().split('\n'))

model = Word2Vec(size=128, window=5, min_count=5, workers=4)
model.build_vocab(corpus,keep_raw_vocab=False)
model.train(corpus)
model.save('w2v')

pprint.pprint(model.most_similar(u'너'))

Above is my source code. It seems like work well. However there are problem with vocabulary key. I want to make korean word2vec which use unicode. For example word 사과 which means apology in english and it's unicode is \\xC0AC\\xACFC If I try to find 사과 in word2vec, key error occur...
Instead of \\xc0ac\\xacfc \\xc0ac and \\xacfc stores separately. What's the reason and how to solve it?

Answer 1

Word2Vec requires text examples that are broken into word-tokens. It appears you are simply providing strings to Word2Vec, so when it iterates over them, it will only be seeing single-characters as words.

Does Korean use spaces to delimit words? If so, break your texts by spaces before handing the list-of-words as a text example to Word2Vec.

If not, you'll need to use some external word-tokenizer (not part of gensim) before passing your sentences to Word2Vec.