src/Gargantext/Text/Terms.hs

   1 {-|
   2 Module      : Gargantext.Text.Ngrams
   3 Description : Ngrams definition and tools
   4 Copyright   : (c) CNRS, 2017 - present
   5 License     : AGPL + CECILL v3
   6 Maintainer  : team@gargantext.org
   7 Stability   : experimental
   8 Portability : POSIX
   9
  10 An @n-gram@ is a contiguous sequence of n items from a given sample of
  11 text. In Gargantext application the items are words, n is a non negative
  12 integer.
  13
  14 Using Latin numerical prefixes, an n-gram of size 1 is referred to as a
  15 "unigram"; size 2 is a "bigram" (or, less commonly, a "digram"); size
  16 3 is a "trigram". English cardinal numbers are sometimes used, e.g.,
  17 "four-gram", "five-gram", and so on.
  18
  19 Source: https://en.wikipedia.org/wiki/Ngrams
  20
  21 TODO
  22 group Ngrams -> Tree
  23 compute occ by node of Tree
  24 group occs according groups
  25
  26 compute cooccurrences
  27 compute graph
  28
  29 -}
  30
  31 {-# LANGUAGE NoImplicitPrelude #-}
  32 {-# LANGUAGE TemplateHaskell   #-}
  33
  34 module Gargantext.Text.Terms
  35   where
  36
  37 import Control.Lens
  38 import Data.Text (Text)
  39 import Data.Traversable
  40
  41 import Gargantext.Prelude
  42 import Gargantext.Core
  43 import Gargantext.Core.Types
  44 import Gargantext.Text.Terms.Multi (multiterms)
  45 import Gargantext.Text.Terms.Mono  (monoTerms)
  46
  47
  48 data TermType lang
  49   = Mono      { _tt_lang :: lang }
  50   | Multi     { _tt_lang :: lang }
  51   | MonoMulti { _tt_lang :: lang }
  52
  53 makeLenses ''TermType
  54
  55 --group :: [Text] -> [Text]
  56 --group = undefined
  57
  58 -- remove Stop Words
  59 -- map (filter (\t -> not . elem t)) $
  60 ------------------------------------------------------------------------
  61 -- | Sugar to extract terms from text (hiddeng mapM from end user).
  62 --extractTerms :: Traversable t => TermType Lang -> t Text -> IO (t [Terms])
  63 extractTerms :: TermType Lang -> [Text] -> IO [[Terms]]
  64 extractTerms termTypeLang = mapM (terms termTypeLang)
  65 ------------------------------------------------------------------------
  66 -- | Terms from Text
  67 -- Mono : mono terms
  68 -- Multi : multi terms
  69 -- MonoMulti : mono and multi
  70 -- TODO : multi terms should exclude mono (intersection is not empty yet)
  71 terms :: TermType Lang -> Text -> IO [Terms]
  72 terms (Mono      lang) txt = pure $ monoTerms lang txt
  73 terms (Multi     lang) txt = multiterms lang txt
  74 terms (MonoMulti lang) txt = terms (Multi lang) txt
  75 -- terms (WithList  list) txt = pure . concat $ extractTermsWithList list txt
  76 ------------------------------------------------------------------------
  77