Strand 5: Computational semantic analysis

Leader : Thierry Charnois, LIPN Paris 13

 

Co-leader: Benoît Crabbé LLF University of Paris

pexels-photo-160107-768x512.jpeg

In the field of computational linguistics, strand 5 emphasizes semantic analysis and its application in various content access tools (e.g. information extraction, indexation, 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 calculation models. 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 parsers that are both robust and of good quality. .

 

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

 

Strand 5 emphasizes the analysis of French, but particular attention is paid to the identification of language-independent methods which make it possible to analyze a wide range of languages for which resources are available. It is indeed a promising approach to equip poorly equipped languages which are used in strands 2 and 3.

 

The strand is interested in written and oral language. The needs for analysis of the spoken language are increasing and one of the challenges to be met is to integrate into the semantic analysis of oral features such as dysfluences to compensate for the low quality of the results obtained on transcriptions of the language. 'oral.

 

The strand has 6 sub-strand:
 

  • Historical and reflective perspective (HP)

  • Corpus constitution and annotation for semantic analysis (IA)

  • Semi-supervised learning for deep parsing (SSL)

  • Specific semantic analysis and processing (SSP)

  • Knowledge Acquisition Methods (KA)

  • Applications: towards enriched access to textual content (APP)

     

Strand 5 operations (2020-2024)

 

SSL5 Shallow semantic analysis

Marie Candito (LLF) - Joseph Leroux (LIPN)

 

SSL6 Semi-supervised Semantic Analysis

Joseph Leroux (LIPN)

 

SSL7 MetaSem: Towards more semantic metagrammars

Djamé Seddah - Eric de la Clergerie

 

SPP6 Word sense disambiguation and sense induction

Lucie Barque - Marie Candito (LLF)

 

SPP7 Computational and Cognitive modeling of the processing of syntactic and semantic information

Benoît Crabbé (LLF)

 

KA2 / APP2 Extraction of semantic relations and applications

Haifa Zargayouna (LIPN)

APP4 RENFO: Expert research on the web through semantic extraction of bio-professional traces from the web

Jorge Garcia Flores (LIPN)

 

APP5 When Aliens meet Predators: Building Context-Enhanced NLP Tools

Djame Seddah

 

APP6 Geo-NN: Geolocalisation in Social Media using Deep Neural Networks

Davide Buscaldi (LIPN)

 

APP7 Legal rule analysis

Adeline Nazarenko (LIPN)

 

HRP2 Graph-based semantic analysis

Adeline Nazarenko (LIPN)

 

KA5 Cross-lingual and multilingual alignment of semantic Resources

Haifa Zargayouna (LIPN)

Strand 5 video presentation