def Analyzer(s):
ls = []
try:
ls = jiagu.sentiment(s)
except:
ls = ['positive', 0.5]
if ls[0] == 'positive':
result = ls[1]
else:
result = 0 - ls[1]
return result
def emotions():
text = []
with open(path + "如何看待4月3日全球新冠确诊人数突破100万_我们需要多久才能控制住疫情_.csv", encoding='gbk',errors='ignore') as file:
for line in file:
line = line.split(',',1) # 只分一次
text.append(line[1])
# print(line[1]) # 数据的查看
# print(len(text)) # 长度100001
with open(path + '如何看待4月3日全球新冠确诊人数突破100万_我们需要多久才能控制住疫情_.txt', 'w', encoding='utf-8') as f:
for i in range(len(text)):
sentiment = jiagu.sentiment(text[i])
f.write(str(sentiment) + '\n')
def senticontent(chapter_id):
contents_sentences = getContentSentence(chapter_id)
if SentiContent.query.filter_by(
sentence_id=contents_sentences[0].id).all() == []:
for i, contents_sentence in enumerate(contents_sentences):
sentiment = jiagu.sentiment(contents_sentence.sentenceseg)
senticontent = SentiContent(senti=sentiment[0],
degree=sentiment[1],
sentence_id=contents_sentence.id)
db.session.add(senticontent)
senticontents = [{
'sentence': contents_sentence.sentenceseg,
'sentiment': senticontent.senti,
'degree': senticontent.degree
} for contents_sentence in contents_sentences
for senticontent in contents_sentence.senti]
return jsonify({'sentiments': senticontents}), 200
def check_comment(url, video):
table_header = [
'appearance', '弹幕模式', '字号', '颜色', '发送时间', '弹幕池', '发送者id', 'rowID',
'content', 'emotion'
]
video_url = url
video_html = open_url(video_url)
danmu_id, video['title'], video['up'] = get_danmu_id(video_html, video_url)
all_list = []
if danmu_id:
danmu_url = 'http://comment.bilibili.com/{}.xml'.format(danmu_id)
danmu_html = open_url(url=danmu_url)
soup = BS(danmu_html, 'html.parser')
all_d = soup.select('d')
for d in all_d:
# 把d标签中P的各个属性分离开
danmu_list = d['p'].split(',')
# d.get_text()是弹幕内容
danmu_list.append(d.get_text())
nature, value = jiagu.sentiment(danmu_list[8])
if nature == 'negative':
value = -value
danmu_list.append(value)
# danmu_list[0] = sec2str(danmu_list[0])
# danmu_list[4] = time.ctime(eval(danmu_list[4]))
all_list.append(danmu_list)
# print(danmu_list)
# all_list.sort()
df = pd.DataFrame(all_list, columns=table_header)
video_df = df.iloc[:, [0, 7, 8, 9]]
bullet_screen_count = video_df.shape[0]
# danmu_emotion = video_df.to_dict(orient='records')
if 'id' in video:
id = 'id'
else:
id = 'aid'
# dict_write(dict_content=danmu_emotion, path='screen_bullet/danmu_emotion/{}.csv'.format(video[id]))
video_df.to_csv('screen_bullet/danmu_emotion/{}.csv'.format(video[id]))
# video['danmu'] = danmu_emotion
video['count'] = bullet_screen_count
return video, danmu_id, video_df.iloc[:, 2]
def text_sentiment_zh():
"""
text sentiment for chinese
---
tags:
- nlp
parameters:
- in: query
name: text
type: string
required: true
default: 你真棒！
description: text content
responses:
200:
description: chinese sentiment response
schema:
type: object
properties:
code:
type: integer
description: status code
sentiment:
type: string
enum: [negative, positive]
probability:
type: integer
"""
text = request.args.get("text")
if text is None:
raise ParameterLostError("sentiment_text")
result = sentiment(text)
return {
"code": 200,
"sentiment": result[0],
"probability": result[1]
}
def nlp_jiagu(self, btn):
text = str(self.lbl.text.strip()).replace("\n", "")
if text[-2:] == "qg":
sentiment = jiagu.sentiment(text[:-2])
self.lbl.text = pprint.pformat(sentiment)
elif text[-2:] == "cq":
keywords = jiagu.keywords(text, 5) # 关键词
self.lbl.text = pprint.pformat(keywords)
elif text[-2:] == "jl":
if "," in self.lbl.text:
docs = self.lbl.text.split(",")
else:
docs = self.lbl.text.split("，")
#print(docs)
cluster = jiagu.text_cluster(docs)
self.lbl.text = pprint.pformat(cluster)
else:
knowledge = jiagu.knowledge(text)
self.lbl.text = pprint.pformat(knowledge)
据观察者网过往报道，2017年我国全国共完成造林736.2万公顷、森林抚育830.2万公顷。其中，天然林资源保护工程完成造林26万公顷，退耕还林工程完成造林91.2万公顷。京津风沙源治理工程完成造林18.5万公顷。三北及长江流域等重点防护林体系工程完成造林99.1万公顷。完成国家储备林建设任务68万公顷。
'''
keywords = jiagu.keywords(text, 5) # 关键词抽取
print(keywords)
summarize = jiagu.summarize(text, 3) # 文本摘要
print(summarize)
# jiagu.findword('input.txt', 'output.txt') # 根据大规模语料，利用信息熵做新词发现。
# 知识图谱关系抽取
text = '姚明1980年9月12日出生于上海市徐汇区，祖籍江苏省苏州市吴江区震泽镇，前中国职业篮球运动员，司职中锋，现任中职联公司董事长兼总经理。'
knowledge = jiagu.knowledge(text)
print(knowledge)
# 情感分析
text = '很讨厌还是个懒鬼'
sentiment = jiagu.sentiment(text)
print(sentiment)
# 文本聚类（需要调参）
docs = [
"百度深度学习中文情感分析工具Senta试用及在线测试", "情感分析是自然语言处理里面一个热门话题",
"AI Challenger 2018 文本挖掘类竞赛相关解决方案及代码汇总", "深度学习实践：从零开始做电影评论文本情感分析",
"BERT相关论文、文章和代码资源汇总", "将不同长度的句子用BERT预训练模型编码，映射到一个固定长度的向量上",
"自然语言处理工具包spaCy介绍", "现在可以快速测试一下spaCy的相关功能，我们以英文数据为例，spaCy目前主要支持英文和德文"
]
cluster = jiagu.text_cluster(docs)
print(cluster)
import jiagu
import pandas as pd
from opencc import OpenCC
# 用來繁轉簡，現在是在要算情緒分數時一筆一筆呼叫轉簡體，效能一定糟透，理論上也許要另外存一個欄位是簡體？
cc = OpenCC('t2s')
txt = pd.read_csv("clean-txt-tokenized.csv")
txt["sentiment"] = 0
for i in range(len(txt)):
txt["sentiment"][i] = jiagu.sentiment(cc.convert(txt["0"][i]))
txt.to_csv("clean-txt-tokenized-sentiment.csv")
async def doDetectSentiment(text: str = Form(...)):
return {'status': True, 'data': jiagu.sentiment(text)[0] == 'positive'}
