inria-00118761, version 1
Conditional Random Fields for XML Trees
Florent Jousse
1, 2, 3Rémi Gilleron 1, 2, 3Isabelle Tellier 1, 2, 3Marc Tommasi 1, 2, 3
Workshop on Mining and Learning in Graphs (2006)
Résumé : We present Conditional Random Fields (XCRFs), a framework for building conditional models to label XML data. XCRFs are Conditional Random Fields over unranked trees (where every node has an unbounded number of children). The maximal cliques of the graph are triangles consisting of a node and two adjacent children. We equip XCRFs with efficient dynamic programming algorithms for inference and parameter estimation. We experiment XCRFs on tree labeling tasks for structured information extraction and schema matching. Experimental results show that labeling with XCRFs is suitable for these problems.
- 1 : MOSTRARE (INRIA Futurs)
- INRIA – CNRS : UMR8022 – Université Lille 1 - Sciences et Technologies : EA3588 – Université Charles de Gaulle - Lille III
- 2 : Laboratoire d'Informatique Fondamentale de Lille (LIFL)
- CNRS : UMR8022 – INRIA – IRCICA – Université Lille 1 - Sciences et Technologies
- 3 : GRAPPA (LIFL)
- CNRS : UMR8022 – Université Charles de Gaulle - Lille III – Université Lille 1 - Sciences et Technologies
- Domaine : Informatique/Apprentissage
- Mots-clés : Graphical Models – Machine Learning – XML – Information Extraction – Labeling – Schema Matching
- inria-00118761, version 1
- http://hal.inria.fr/inria-00118761
- oai:hal.inria.fr:inria-00118761
- Contributeur : Marc Tommasi
- Soumis le : Mercredi 6 Décembre 2006, 13:25:05
- Dernière modification le : Lundi 14 Décembre 2009, 22:34:59
Documents associés
Exporter