This work describes a model that allows qualifying a keyword in a scientific article, considering how much it is used in the text and according to some weightings established by members of the Community of Engineering Journal Editors in Colombia. For this, the reader is contextualized in necessary concepts such as metadata, Semantic Web, Big Data, and then, the components of the model are formulated along with two algorithms used in the model presented. Finally, there is an analysis of the application of the model in the Facultad de Ingenieria Journal of the Universidad Pedagógica y Tecnológica de Colombia.