In this paper, we present a proposal for searching, retrieving and recommending documents, based on the similarity of their content, using an ontology of the type “Friend of a Friend” that establishes, if it is the case, the transitivity of similarity between the document selected by the user and the other documents in the repository. Document similarity is calculated as the cosine of the angle between the vectors of terms that represent them. The Vector Space Model is used for representing the documents and the weight of the terms is obtained by using the TF-IDF metric. The architecture that implements the proposal is presented as well as experimental results.
