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 OverloadedStrings #-}
  33 {-# LANGUAGE TemplateHaskell   #-}
  34
  35 module Gargantext.Text.Terms
  36   where
  37
  38 import Control.Lens
  39 import Data.Text (Text)
  40 import Data.Traversable
  41 import GHC.Base (String)
  42
  43 import Gargantext.Prelude
  44 import Gargantext.Core
  45 import Gargantext.Core.Types
  46 import Gargantext.Text.Terms.Multi (multiterms)
  47 import Gargantext.Text.Terms.Mono  (monoTerms)
  48 import Gargantext.Text.Terms.Mono.Stem (stem)
  49
  50 import qualified Data.Set  as Set
  51 import qualified Data.List as List
  52 import qualified Data.Text as Text
  53 import Gargantext.Text (sentences)
  54 import Gargantext.Text.Terms.Mono.Token.En (tokenize)
  55 import Gargantext.Text.Terms.Eleve (mainEleveWith, Tries, Token, buildTries, toToken)
  56
  57 data TermType lang
  58   = Mono      { _tt_lang :: lang }
  59   | Multi     { _tt_lang :: lang }
  60   | MonoMulti { _tt_lang :: lang }
  61   | Unsupervised { _tt_lang  :: lang
  62                  , _tt_windoSize  :: Int
  63                  , _tt_ngramsSize :: Int
  64                  , _tt_model :: Maybe (Tries Token ())
  65   }
  66 makeLenses ''TermType
  67
  68 --group :: [Text] -> [Text]
  69 --group = undefined
  70
  71 -- remove Stop Words
  72 -- map (filter (\t -> not . elem t)) $
  73 ------------------------------------------------------------------------
  74 -- | Sugar to extract terms from text (hiddeng mapM from end user).
  75 --extractTerms :: Traversable t => TermType Lang -> t Text -> IO (t [Terms])
  76 extractTerms :: TermType Lang -> [Text] -> IO [[Terms]]
  77
  78 extractTerms (Unsupervised l n s m) xs = mapM (terms (Unsupervised l n s (Just m'))) xs
  79   where
  80     m' = case m of
  81       Just m''-> m''
  82       Nothing -> newTries n (Text.intercalate " " xs)
  83
  84 extractTerms termTypeLang xs = mapM (terms termTypeLang) xs
  85
  86 ------------------------------------------------------------------------
  87 -- | Terms from Text
  88 -- Mono : mono terms
  89 -- Multi : multi terms
  90 -- MonoMulti : mono and multi
  91 -- TODO : multi terms should exclude mono (intersection is not empty yet)
  92 terms :: TermType Lang -> Text -> IO [Terms]
  93 terms (Mono      lang) txt = pure $ monoTerms lang txt
  94 terms (Multi     lang) txt = multiterms lang txt
  95 terms (MonoMulti lang) txt = terms (Multi lang) txt
  96 terms (Unsupervised lang n s m) txt = termsUnsupervised (Unsupervised lang n s (Just m')) txt
  97   where
  98     m' = maybe (newTries n txt) identity m
  99 -- terms (WithList  list) txt = pure . concat $ extractTermsWithList list txt
 100 ------------------------------------------------------------------------
 101
 102 text2term :: Lang -> [Text] -> Terms
 103 text2term _ [] = Terms [] Set.empty
 104 text2term lang txt = Terms txt (Set.fromList $ map (stem lang) txt)
 105
 106 isPunctuation :: Text -> Bool
 107 isPunctuation x = List.elem x $  (Text.pack . pure)
 108                              <$> ("!?(),;." :: String)
 109
 110 -- | Unsupervised ngrams extraction
 111 -- language agnostic extraction
 112 -- TODO: remove IO
 113 -- TODO: newtype BlockText
 114
 115 type WindowSize = Int
 116 type MinNgramSize = Int
 117
 118 termsUnsupervised :: TermType Lang -> Text -> IO [Terms]
 119 termsUnsupervised (Unsupervised l n s m) =
 120                pure
 121              . map (text2term l)
 122              . List.nub
 123              . (List.filter (\l' -> List.length l' >= s))
 124              . List.concat
 125              . mainEleveWith (maybe (panic "no model") identity m) n
 126              . uniText
 127 termsUnsupervised _ = undefined
 128
 129 newTries :: Int -> Text -> Tries Token ()
 130 newTries n t = buildTries n (fmap toToken $ uniText t)
 131
 132 -- | TODO removing long terms > 24
 133 uniText :: Text -> [[Text]]
 134 uniText = map (List.filter (not . isPunctuation))
 135         . map tokenize
 136         . sentences       -- | TODO get sentences according to lang
 137         . Text.toLower
 138