如何在基于频率的情感分析中找到与名词短语配对的情感?

时间:2019-10-16 05:51:54

标签: python sentiment-analysis

我有一个基于情感分析频率的项目。我已经按照教授的步骤进行了操作:

  1. POS标签评论(检查)
  2. 提取名词短语(检查)
  3. 计算名词短语频率(检查)
  4. 将名词短语频率从低到高排序(检查)
  5. 过滤低频名词短语(检查)的阈值
  6. 找到将名词短语配对的情感(尚未)
  7. 查找不频繁(尚未)
  8. 查找精度(尚未)

接下来的部分是寻找我的名词短语的情感/观点。我还没有尝试过任何东西,因为我仍在互联网上搜索。但是当我写这篇文章时,我还没有找到它们。

这些是高频名词短语(我使用阈值= 2):

("i 'd", 3)
('powershot g3', 3)
('digital photography', 3)
('digital pictures', 3)
("n't need", 3)
('shutter speed', 3)
('raw images', 3)
('image quality', 3)
('compact flash', 3)
('view finder', 3)
('megapixel camera', 3)
('nikon coolpix', 3)
('external flash', 4)
('lcd screen', 4)
('lens cap', 4)
('great pictures', 4)
('great camera', 4)
('metz flash', 4)
('raw image', 4)
('auto mode', 5)
('picture quality', 7)
('battery life', 9)
('canon g3', 11)
('digital cameras', 11)
('digital camera', 12)
("i 'm", 15)
("i 've", 19)

以下是一些评论:

just a little overview , powershot g3 is the flagship of canon's powershot series and its an slr-like camera , its 4 megapixel and ( alsmost ) full manual control gives the pictures a touch of brilliance

despite this minor disappointment , i highly recommend the canon g3 anyone who is serious about digital photography

recent price drops have made the g3 the best bargain in digital cameras currently available

the olympus is a bit clumsy-looking and the user-interface not as friendly as the canon , but one of the features that sold me on the g3 was the battery life - no other camera out there gives you the type of battery life as the canon g3

(注意:我...实际上是使用textblob的,所以我可以更轻松地获得名词短语,但是我还是会在评论中加上POS标签,以防万一。)

0 个答案:

没有答案