src/Gargantext/Pipeline.hs

   1 {-|
   2 Module      : Gargantext.Pipeline
   3 Description : Server API
   4 Copyright   : (c) CNRS, 2017-Present
   5 License     : AGPL + CECILL v3
   6 Maintainer  : team@gargantext.org
   7 Stability   : experimental
   8 Portability : POSIX
   9
  10 -}
  11
  12 {-# OPTIONS_GHC -fno-warn-name-shadowing #-}
  13 {-# LANGUAGE NoImplicitPrelude           #-}
  14
  15 module Gargantext.Pipeline
  16   where
  17
  18 import Data.Text.IO (readFile)
  19
  20 ----------------------------------------------
  21 ----------------------------------------------
  22 import Gargantext.Core
  23 import Gargantext.Prelude
  24
  25 import Gargantext.Viz.Graph.Index (score)
  26 import Gargantext.Viz.Graph.Distances.Matrice (distributional)
  27 import Gargantext.Text.Metrics.Occurrences
  28 import Gargantext.Text.Terms
  29 import Gargantext.Text.Context
  30
  31
  32 pipeline path = do
  33   -- Text  <- IO Text <- FilePath
  34   text     <- readFile path
  35   let contexts = splitBy (Sentences 3) text
  36   myterms <- extractTerms Multi FR contexts
  37   -- TODO    filter (\t -> not . elem t stopList) myterms
  38   -- TODO    groupBy (Stem | GroupList)
  39   let myCooc = removeApax $ cooc myterms
  40   -- Cooc -> Matrix
  41   pure $ score distributional myCooc
  42   -- Matrix -> Clustering -> Graph -> JSON
  43