Strand 5 : Computational semantic analysis (manager : A. Nazarenko, LIPN Paris 13)

In the field of computational linguistics, Axis 5 focuses on semantic analysis and its application in various content access tools (eg information extraction, indexing, summary), which may be based on different semantic representations. Models, methods and calculations and – in some cases – tools have been proposed to describe and model different semantic phenomena but these phenomena have generally been considered in isolation and with heterogeneous models of computation. The challenge is to evaluate these different semantic models, to operationalize them and to integrate them to produce rich, covering and coherent semantic analyzes, in the same way, that we have been able to produce robust and good quality parsers.

The goal is to intensify the research on the integration of semantic models and methods by the collaboration of different teams of computational linguistics (Alpage, Lattice, LIPN, LPP-P3) and the comparison of their heterogeneous results on parsing textual analysis, analysis and corpus processing, inferential reasoning based on texts, acquisition of knowledge from texts, machine learning and tools for accessing textual content.

Strand 5 focuses on the analysis of French, but particular attention is paid to the identification of language-independent methods for analyzing a wide range of languages for which resources are available. It is indeed a promising approach to equip languages with little tools that are used in strand 2 and 3.

The strand focuses on the written and oral language. The need for analysis of the spoken language is increasing and one of the challenges to be met is to integrate in the semantic analysis of oral features such as the dysfluences to compensate for the poor quality of the results obtained on transcripts of the spoken language. ‘oral.


The strand has 6 sub-axes :

·     Historical and Reflective Perspective (HP)

·     Specific analyzes on semantic processing (SSP)

·     Knowledge acquisition methods (KA)

·     Semi-supervised learning (SSL)

·     Integrative approaches (IA)

·       Applications: towards enhanced access to textual content (APP)

Lists of current search operations :

 

Historical and Reflective Perspective (HP)

  • Theory and meaning and TAL (HP1, resp. J. Léon)

Specific analyzes on semantic processing (SSP)

  • Specific oral characteristics for semantic processing (SSP2, resp. M. Adda-Decker)
  • Filtering non-discursive uses of connectors (SSP3, resp. L. Danlos)
  • Deep syntactic annotation (SSP4, resp. M. Candito)

Knowledge acquisition methods (KA)

  • Automatic induction of representative lexico-grammatical patterns from texts (KA2, resp. T. Poibeau)

  • Acquiring domain knowledge from texts (KA3, resp. A. Nazarenko)

Semi-supervised learning (SSL)

  • Contribution of parsing to information extraction: domain adaptation and multi-objective learning (SSL3, resp. J. Le Roux)

Integrative approaches (IA)

  • Annotation platform (IA1)

Applications: towards enriched access to textual content (APP)

  • Design and development of new methods of accessing textual content (APP2, resp. H. Zargayouna)