The extraction of the information from database systems requires the formulation of queries using database query languages, such as Structured Query Language (SQL). This formulation needs the knowledge of the model and the structure of the database. However, non-expert users cannot write such queries. This is why a lot of works have been developed to request the database in natural language. Historically, most of these works were carried out for English language and they were designed for a specific database model. Some of them function independently of database domain. For the French language, all existing contributions are dependent on database domain and model. The aim of this paper is to present a model of an intelligent interface to query databases using the French language. A simulation model was established through the using of linguistic operations and machine learning. The results indicate that the proposed interface functions independently of the database domain and model (relational and XML) and it can translate a very important number of French natural language query into a database query. The findings of this research may serve to improve the interaction between non-expert French users and databases.