The NLP team of LIRMM currently  works on lexical
disambiguation and thematic  text analysis \cite{Laf2001a}.
We built a system, with automated learning capabilities, based on
conceptual vectors for meaning representation.
Vectors are supposed to encode  \emph{ideas} associated to words or
expressions. In the framework of knowledge and lexical meaning representation,
 we devise some  conceptual
vectors based strategies to automatically construct hierarchical taxonomies and validate (or invalidate)
hyperonymy (or superordinate) relations among terms.
Conceptual vectors are used through the thematic distance for decision making and link quality assessment.