The subject Brazil was analysed within the context of the French data base DocThéses, comprising the years 1969 up to 1999. The data minig technique was used to obtain intelligence and infer knowledge. The software used to do the cleaning of the base DocThèses was Infotrans; and for the preparation of the data was Dataview. The results of the analysis were illustrated by making use of the assumptions of the Zipf Law, on bibliometrics, classifying the information in trivial information; of interest; and "noise", according to the distribution of frequency. The conclusion is that the Data Mining technique associated with specialist software is a powerful ally for the competitive intelligence applied on all levels of the decision making process, including the macro level. It can enhance the consolidation, investment and development of actions and policies.
Data mining; Bibliometrics; Bibliometrical analysis; French thesis; Brazil; Knowledge discovery; Data bases; Zipf Law
