-1

Are there NLP algorithms dealing with detecting the repeating patterns in a a list of texts from which a topic keywords and other associative keywords can be derived?

I will show it as an example: You have a search query "vegan food for something health" (where something is a part of body you need an advice about).

The search engine will return a list of articles.

The algorithm will search for patterns in these articles. E.g. it notices that 80 % of them have a paragraph with at least 4 multiple instances of a word orange, similarly carrot, apples, cucumbers.

So it will give you an outline (textual mindmap)

  • orange
  • carrot --> vitamin A
  • apple
  • banana --> vitamin B
  • run a lot

Once I watched a video about semantic web on youtube and know that Tim Berners-Lee talked about something similar, but I have lost the link. Could you keyword me to that direction again?

Stanislav Kralin
  • 11,070
  • 4
  • 35
  • 58
xralf
  • 3,312
  • 45
  • 129
  • 200

1 Answers1

1

Probably you are looking for word2vec -- described patterns can be described in terms of distance between words.

dveim
  • 3,381
  • 2
  • 21
  • 31