src/Gargantext/Text/Terms/Multi.hs

   1 {-|
   2 Module      : Gargantext.Text.Terms.Multi
   3 Description : Multi Terms module
   4 Copyright   : (c) CNRS, 2017 - present
   5 License     : AGPL + CECILL v3
   6 Maintainer  : team@gargantext.org
   7 Stability   : experimental
   8 Portability : POSIX
   9
  10 Multi-terms are ngrams where n > 1.
  11
  12 -}
  13
  14 {-# LANGUAGE NoImplicitPrelude #-}
  15
  16 module Gargantext.Text.Terms.Multi (multiterms)
  17   where
  18
  19 import Data.Text hiding (map, group, filter, concat)
  20 import Data.List (concat)
  21
  22 import Gargantext.Prelude
  23 import Gargantext.Core (Lang(..))
  24 import Gargantext.Core.Types
  25
  26 import Gargantext.Text.Terms.Multi.PosTagging
  27 import qualified Gargantext.Text.Terms.Multi.Lang.En as En
  28 import qualified Gargantext.Text.Terms.Multi.Lang.Fr as Fr
  29
  30 multiterms :: Lang -> Text -> IO [Terms]
  31 multiterms lang txt = concat
  32                    <$> map (map tokenTag2terms)
  33                    <$> map (filter (\t -> _my_token_pos t == Just NP))
  34                    <$> tokenTags lang txt
  35
  36 tokenTag2terms :: TokenTag -> Terms
  37 tokenTag2terms (TokenTag w t _ _) =  Terms w t
  38
  39 tokenTags :: Lang -> Text -> IO [[TokenTag]]
  40 tokenTags lang s = map (group lang) <$> tokenTags' lang s
  41
  42
  43 tokenTags' :: Lang -> Text -> IO [[TokenTag]]
  44 tokenTags' lang t =  map tokens2tokensTags
  45                      <$> map _sentenceTokens
  46                      <$> _sentences
  47                      <$> corenlp lang t
  48
  49 ---- | This function analyses and groups (or not) ngrams according to
  50 ----   specific grammars of each language.
  51 group :: Lang -> [TokenTag] -> [TokenTag]
  52 group EN = En.group
  53 group FR = Fr.group
  54