A star-graph is a conceptual graph that contains a single relation, with some concepts linked to it. They are elementary pieces of information describing combinations of concepts. We use star-graphs as descriptors — or index terms — for image content representation. This allows for relational indexing and expression of complex user needs, in comparison to classical text retrieval, where simple keywords are generally used as document descriptors. In classical text retrieval, the keywords are weighted to give emphasis to good document descriptors and discriminators where the most popular weighting schemes are based on variations of tf.idf. In this paper, we present an extension of tf.idf, introducing a new weighting scheme suited for star-graphs. This weighting scheme is based on a local analysis of star-graphs indexing a document and a global analysis of star-graphs across the whole collection. We show and discuss some preliminary results evaluating the performance of this weighting scheme applied to image retrieval.