1

I have a list f keywords i used TextRank algorithme, and i have list of all words tokinized ,i want to combine kaywords into a multi-word keyword depending on their position in the list of all words tokenized with java. For exemple :

Keywords :[ management ,activities ,such ,accounting ,risk ,project ,performance]

list of all words tokenized :

[ "A", "cloud", "service", "for", "managing", "business", "activities", "such", "accounting",",", "procurement", "project", "management",",","risk", "management"," performance","."]

Result of multi-Keywords :

[ "risk management" ,"activities such accounting","project","management performance"]
Eritrean
  • 15,851
  • 3
  • 22
  • 28
Sarah
  • 11
  • 1

1 Answers1

-1

The results that you want ("multi-word keyword") are phrases. We have a Python implementation, for multiple textrank algorithms, which handles that by default – at https://github.com/DerwenAI/pytextrank/

Paco
  • 602
  • 1
  • 9
  • 19
  • Please avoid posting link-only answers, see https://meta.stackexchange.com/questions/225370/your-answer-is-in-another-castle-when-is-an-answer-not-an-answer – Progman Mar 05 '22 at 09:05
  • Thank you ,but i'm searching for java implementation , i already implemented the algorithm to extract keywords, but i didn't understand or find any solution how to combine keywords to phrases. – Sarah Mar 05 '22 at 14:40
  • This is not a link-only answer. The question about NLP is answered by the need for key phrase extraction. Implementation of that within the _textgraph_ family of algorithms is much longer than would fit well here, although there are multiple open source projects which show this kind of approach. – Paco Mar 06 '22 at 18:01