Natural Language Processing fran├žais

The work done on the text analysis topic are based on LIMA, the LIST Multilingual Analyzer, a powerful and modular tool able to syntactically parse text and extract named entities in 9 languages (English, French, German, Spanish, Italian, Russian, Arabic, Chinese and Hungarian).


This analyzer supports work on various fields where Natural Language Processing is implied.

For example, it is used in our multimedia and crosslingual search engine and it is also the basis of research efforts in classification and clustering. It supports some development in semantic analysis, including the semi-automatic building of ontologies, the acquisition of various semantic resources (terminologies, semantic map, FrameNet and WordNet-like resources) and the development of semantic annotators for Word Sense Disambiguation or Semantic Role Labeling. Finally, LIMA is used in applications like automatic summarization and Question and Answer systems.