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 import qualified Data.Map.Strict as M
  20 ----------------------------------------------
  21 import Gargantext.Core (Lang(FR))
  22 import Gargantext.Prelude
  23
  24 import Gargantext.Viz.Graph.Index (score, createIndices, toIndex)
  25 import Gargantext.Viz.Graph.Distances.Matrice (conditional)
  26 import Gargantext.Text.Metrics.Occurrences (cooc, removeApax)
  27 import Gargantext.Text.Terms (TermType(Multi, Mono), extractTerms)
  28 import Gargantext.Text.Context (splitBy, SplitContext(Sentences))
  29
  30 import Data.Graph.Clustering.Louvain.CplusPlus (cLouvain)
  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
  38   -- TODO    filter (\t -> not . elem t stopList) myterms
  39   -- TODO    groupBy (Stem | GroupList)
  40
  41   let myCooc = removeApax $ cooc myterms
  42   -- Cooc -> Matrix
  43   let theScores = M.take 350 $ M.filter (>0) $ score conditional myCooc
  44   let (ti, _) = createIndices theScores
  45 --
  46 ----  -- Matrix -> Clustering -> Graph -> JSON
  47 ----  pure $ bestpartition False $ map2graph $ toIndex ti theScores
  48   partitions <- cLouvain $ toIndex ti theScores
  49   pure partitions
  50