text - 使用 sklearn 或现有模块对文本进行聚类，让每个聚类属于多标签

Question

我有这样的数据

  [
  'The actor in New York. The art of Static.',
  'The actor in New York. Sword.',
  'The actor in New York. Handsome Jonny Deep.',
  'France, the greast tower. In Las Vegas, the great hotel.',
  'Empire State Building. This night is great!',
  'Empire State Building wow! This night is so so so so great!',
  'The light of life. Make you think of the future ans the love.',
  'The light of street. Make you consider that where is my road tomorrow?',
  'A scrolled cat palying on the ground. cute',
  'A surprised woman in a dark raining day.',
  'The sunshine makes the cat wants to sleep.',
  'Three in one image that a cat is smelling the flower.',
  'A woman with smile is looking at the falling leaves.',
  'A headshot of a fat squirrel.',
  'A headshot of a yellow cat.',
  'A cat runs on the snow and jumps very high.',
  'A dog grovals on the falling leaves and looks at camera.',
  'One cat stares at the camera, another is looking around.',
  'A dog with long tongue plays on the grass.'
  'A sad dog wants to see the world outside the house.'
  ]

我想对它们进行聚类并用多个标签标记它们中的每一个。可以使用现有工具吗？我没有足够的时间自己实施。太感谢了！！

Answer 1

您可以尝试使用gensim：http: //radimrehurek.com/gensim/

您可以使用 LDA，这种方法可以为您的每个文档（短语）分配标签（主题）。然后，您可以使用一种方法根据主题的分布对它们进行聚类——例如，可以使用 K-Means。

请看一下：http ://radimrehurek.com/gensim/models/ldamodel.html

然后你使用 sklearn 中的 K-Means 方法。