Trace Corpus et Apprentissage

1
Trace Corpus et Apprentissage
Christian Champendal Champend_at_unice.fr Ecole
Polytechnique Université Nice Sophia-Antipolis
2
Constitution de Corpus

• Tri documentaire

Doc1
Docn
3
Corpus
Index ..\..\Elec2\Manip1\Ztext.exe
Docn
4
Extraction de Résultats

• Lexique Syntaxe

a1b1 a2bn b2r1 bnrn a1b2rn b2r1an
a1ltp1gtb2 b2ltp2gtan cn cnltpngtbn rn
a1 a2 b1 bn r1 rn
5
Séries Lexicales
6
Formule Prédicative

• C0 Sujet (S) ltPredicatgtC1Complément(Od)
  C2Complément(Od/Oi) CircnCirconstanciels
  (Adjct) 
• _____________________________________
• C0 ltUsegtC1 Circ1ASA Circ2TO Circ3IN
• The algorithm uses two neighbouring pixels, one
  left and the other top, as a pioneering block to
  search for the best matched blocks inside a
  pre-defined window.
• C0 ltFiltergtC1 Circ1FROM Circ2 IN/OVER
  Circ3 TO Circ4 BY USE OF
• In WI, pitch-cycle waveforms are filtered in the
  evolution domain to decompose the signal into
  two waveform surfaces, one characterising voiced
  speech and a second representing unvoiced speech.

7
Agrégats
8
Représentation
Sémantique Connaissances Concept ? Ontologi
e Expert Rôle thématique