def sentence_score(sentence):
final_score = []
cuted_review = tp.cut_sentence(sentence) #cut sentence into subsentences
# for w in cuted_review:
# print w
for sent in cuted_review:
seg_sent = tp.segmentation(sent) # segment words
seg_sent = tp.del_stopwords(seg_sent)[:]
# for w in seg_sent:
# print w
i = 0 # current location
s = 0 # emotion word location
poscount = 0 # positive word score
negcount = 0 # negative word score
for word in seg_sent:
# print word
if word in posdict:
# print word
poscount += 1
for w in seg_sent[s:i]:
# print w
poscount = match(w, poscount)
# print poscount
s = i + 1
elif word in negdict:
negcount += 1
for w in seg_sent[s:i]:
negcount = match(w, negcount)
s = i + 1
# if ! ！, which means coming to end of sentence
elif word == "！".decode("utf-8") or word == "!".decode('utf-8'):
for w2 in seg_sent[::-1]:
if w2 in posdict:
poscount += 2
break
elif w2 in negdict:
negcount += 2
break
i += 1
final_score.append(transform_to_positive_num(poscount, negcount)) # final process
pos_result, neg_result = 0, 0
for res1, res2 in final_score: # 每个分句循环累加
pos_result += res1
neg_result += res2
#print pos_result, neg_result
result = pos_result - neg_result # final score
return result
def single_review_sentiment_score(comment_sent):
single_review_senti_score = []
cuted_review = tp.cut_sentence(comment_sent) # 句子切分，单独对每个句子进行分析
for sent in cuted_review:
seg_sent = tp.segmentation(sent) # 分词
seg_sent = tp.del_stopwords(seg_sent)[:]
#for w in seg_sent:
# print w,
i = 0 # 记录扫描到的词的位置
s = 0 # 记录情感词的位置
poscount = 0 # 记录该分句中的积极情感得分
negcount = 0 # 记录该分句中的消极情感得分
for word in seg_sent: # 逐词分析
#print word
if word in posdict: # 如果是积极情感词
#print "posword:", word
poscount += 1 # 积极得分+1
for w in seg_sent[s:i]:
poscount = match(w, poscount)
#print "poscount:", poscount
s = i + 1 # 记录情感词的位置变化
elif word in negdict: # 如果是消极情感词
#print "negword:", word
negcount += 1
for w in seg_sent[s:i]:
negcount = match(w, negcount)
#print "negcount:", negcount
s = i + 1
# 如果是感叹号，表示已经到本句句尾
elif word == "！" or word == "!":
for w2 in seg_sent[::-1]: # 倒序扫描感叹号前的情感词，发现后权值+2，然后退出循环
if w2 in posdict:
poscount += 2
break
elif w2 in negdict:
negcount += 2
break
i += 1
#print "poscount,negcount", poscount, negcount
single_review_senti_score.append(
transform_to_positive_num(poscount, negcount)) # 对得分做最后处理
pos_result, neg_result = 0, 0 # 分别记录积极情感总得分和消极情感总得分
for res1, res2 in single_review_senti_score: # 每个分句循环累加
pos_result += res1
neg_result += res2
#print pos_result, neg_result
result = pos_result - neg_result # 该条评论情感的最终得分
result = round(result, 1)
return result
def single_review_sentiment_score(weibo_sent):
single_review_senti_score = []
cuted_review = tp.cut_sentence(weibo_sent) # 句子切分，单独对每个句子进行分析
for sent in cuted_review:
seg_sent = tp.segmentation(sent) # 分词
seg_sent = tp.del_stopwords(seg_sent)[:]
#for w in seg_sent:
# print w,
i = 0 # 记录扫描到的词的位置
s = 0 # 记录情感词的位置
poscount = 0 # 记录该分句中的积极情感得分
negcount = 0 # 记录该分句中的消极情感得分
for word in seg_sent: # 逐词分析
#print word
if word in posdict: # 如果是积极情感词
#print "posword:", word
poscount += 1 # 积极得分+1
for w in seg_sent[s:i]:
poscount = match(w, poscount)
#print "poscount:", poscount
s = i + 1 # 记录情感词的位置变化
elif word in negdict: # 如果是消极情感词
#print "negword:", word
negcount += 1
for w in seg_sent[s:i]:
negcount = match(w, negcount)
#print "negcount:", negcount
s = i + 1
# 如果是感叹号，表示已经到本句句尾
elif word == "！".decode("utf-8") or word == "!".decode('utf-8'):
for w2 in seg_sent[::-1]: # 倒序扫描感叹号前的情感词，发现后权值+2，然后退出循环
if w2 in posdict:
poscount += 2
break
elif w2 in negdict:
negcount += 2
break
i += 1
#print "poscount,negcount", poscount, negcount
single_review_senti_score.append(transform_to_positive_num(poscount, negcount)) # 对得分做最后处理
pos_result, neg_result = 0, 0 # 分别记录积极情感总得分和消极情感总得分
for res1, res2 in single_review_senti_score: # 每个分句循环累加
pos_result += res1
neg_result += res2
#print pos_result, neg_result
result = pos_result - neg_result # 该条微博情感的最终得分
result = round(result, 1)
return result
def single_review_sentiment_score(pinglun_sent):
total = jieba.analyse.extract_tags(pinglun_sent,
topK=20,
withWeight=True,
allowPOS=()) #找关键词
'''
for each in total:
try:
tmp = Dict.objects.get(word=each[0])
if tmp.type == 'pos':
print(each, tmp.type)
elif tmp.type == 'neg':
print(each, tmp.type)
except:
print(each, '没找到')
'''
seg_sent = tp.segmentation(pinglun_sent) # 分词
i = 0 # 记录扫描到的词的位置
s = 0 # 记录情感词的位置
pos_count = 0 # 记录该分句中的积极情感得分
neg_count = 0 # 记录该分句中的消极情感得分
for each in seg_sent: # 逐词分析
try:
tmp = Dict.objects.get(word=each[0])
if tmp.type == 'pos':
print(each, tmp.type)
for w in total:
if each == w[0]:
pos_count += w[1]
pos_count += 1
for w in seg_sent[s:i]:
pos_count = match(w, pos_count)
s = i + 1 # 记录情感词的位置变化
elif tmp.type == 'neg': # 如果是消极情感词
print(each, tmp.type)
for w in total:
if each == w[0]:
neg_count += w[1]
neg_count += 1
for w in seg_sent[s:i]:
neg_count = match(w, neg_count)
s = i + 1 # 记录情感词的位置变化
except:
print(each, '没找到')
i += 1
total_count = len(total)
print(pos_count, neg_count, total_count,
float(pos_count - neg_count) / total_count)
return float(pos_count - neg_count) / total_count
def calculate_score(self, content):
total_score = 0.0
cut_contexts = tp.cut_sentence(content)
isKeyWord = False
# 对应每句话
# print 'content: ', content
for cut_context in cut_contexts:
words = tp.segmentation(cut_context)
# 去停用词
words = self.del_stop_words(words)
score = 0.0
prefix = 1.0
for word in words:
if word in self.inverse_dict:
prefix *= self.weight_inverse
# print "inverse word:", word, self.weight_inverse
elif word in self.ish_dict:
prefix *= self.weight_ish
# print "ish word", word, self.weight_ish
elif word in self.more_dict:
prefix *= self.weight_more
# print "more word:", word, self.weight_more
elif word in self.very_dict:
prefix *= self.weight_very
# print "very word:", word, self.weight_very
elif word in self.most_dict:
prefix *= self.weight_most
# print "most word", word, self.weight_most
# if len(word) >= 2:
else:
result, flag = self.word_search(word)
if flag:
score += result
if word in self.key_words:
isKeyWord = True
# print 'key',word
# print 'word:', word, score
score_final = prefix * score
# print 'sentence:', cut_context, score_final, prefix
# if (score_final > -0.001) & (score_final < 0.001):
# score_final = self.SnowNLP_analyze(cut_context)
# print "final_socre",score_final
total_score += score_final
# print 'total score:', total_score
if isKeyWord & (total_score < -0.01):
total_score += self.weight_key_word
return total_score
def single_review_sentiment_score(weibo_sent):
single_review_senti_score = []
cuted_review = tp.cut_sentence(weibo_sent)
for sent in cuted_review:
seg_sent = tp.segmentation(sent)
'''
if seg_sent[0]=='@':
continue
'''
i = 0 # 记录扫描到的词的位置
s = 0 # 记录情感词的位置
poscount = 0 # 记录该分句中的积极情感得分
negcount = 0 # 记录该分句中的消极情感得分
for word in seg_sent:
if word in posdict:
poscount += 1
for w in seg_sent[s:i]:
poscount = match(w, poscount)
s = i + 1
elif word in negdict:
negcount += 1
for w in seg_sent[s:i]:
negcount = match(w, negcount)
s = i + 1
elif word == "吗".encode('utf-8').decode("utf-8"):
for w2 in seg_sent[::-1]:
if w2 in posdict:
poscount = poscount * -1
break
elif w2 in negdict:
negcount = negcount * -1
break
elif word == "？".encode('utf-8').decode(
"utf-8") or word == "?".encode('utf-8').decode("utf-8"):
if i + 1 < len(seg_sent):
for w2 in seg_sent[i + 1:]:
if w2 == '？'.encode('utf-8').decode(
"utf-8") or w2 == '?'.encode('utf-8').decode(
"utf-8"):
if negcount > poscount:
negcount = negcount + 1
else:
poscount = poscount + 1
if negcount == 0 and poscount == 0:
negcount = negcount + 1
elif word == "！".encode('utf-8').decode(
"utf-8") or word == "!".encode('utf-8').decode("utf-8"):
m = 0
for w2 in seg_sent[::-1]:
if w2 == "！".encode('utf-8').decode(
"utf-8") or word == "!".encode('utf-8').decode(
"utf-8"):
m = m + 1
if w2 in posdict:
poscount = poscount + 1.5 * m
break
elif w2 in negdict:
negcount = negcount + 1.5 * m
break
i += 1
single_review_senti_score.append(
transform_to_positive_num(poscount, negcount))
'''
poscount, negcount=transform_to_positive_num(poscount, negcount)
single_review_senti_score.append(poscount-negcount)
single_review_senti_score=np.array(single_review_senti_score)
if sum(single_review_senti_score)==0:
result=0
else:
result=np.mean(single_review_senti_score[single_review_senti_score!=0])
return result
'''
pos_result, neg_result = 0, 0
for res1, res2 in single_review_senti_score:
pos_result += res1
neg_result += res2
result = pos_result - neg_result
result = round(result, 2)
if result > 10:
result = 10
elif result < -10:
result = -10
return result
import text_process as tp
import jieba
jieba.load_userdict(
'E:\postgraduate\\no_space_environment\category\pycharm\pycharm_file_location\\thesis\新闻文本处理\论文词典法CSDN\Sentiment_dict\emotion_dict\pos_all_dict.txt'
)
jieba.load_userdict(
'E:\postgraduate\\no_space_environment\category\pycharm\pycharm_file_location\\thesis\新闻文本处理\论文词典法CSDN\Sentiment_dict\emotion_dict\\neg_all_dict.txt'
)
jieba.load_userdict(
"E:\postgraduate\\no_space_environment\category\pycharm\pycharm_file_location\\thesis\新闻文本处理\论文词典法CSDN\Sentiment_dict\degree_dict\insufficiently_inverse.txt"
)
news_sent = '我不是很看好这支股票会上涨。'
# cuted_review = tp.cut_sentence(news_sent) # 句子切分，单独对每个句子进行分析
cuted_review = list(jieba.cut(news_sent))
print(cuted_review)
for sent in cuted_review:
seg_sent = tp.segmentation(sent) # 分词
seg_sent = tp.del_stopwords(seg_sent)[:]
print(seg_sent)
