As a tester / developer / business analyst / customer

I want to see which area is the most "infected" (where are the most of the bugs)

And I want to get rid of duplicate bugs

And i want to see bugs connected to their stories

And I want to see stories grouped together by their subject to follow changes easily

So please implement Natural Language Processing to group / cluster issues based on their title, type, description, labels and comments. 

This would also improve issue search itself (filtering)

Most of the multinational companies use Jira in English, so English IT dictionaries and/or existing Jiras would be good as teaching data.

This would be a major improvement, so may be an extra service for some additional fee paid.