Bibliographic data : a different analysis perspective

A bibliographic record, related to a product, is composed by different information: authors, year, source, publisher, keywords, abstract, citations and so on. Citations usually have a central role in bibliometric analysis. The study of textual information could be a different analysis perspective. The idea is that documents are mixture of latent topics, where a topic is a probability distribution over words. In this paper we try to show how the scientific productivity of a research group can be described using topic models. Moreover, for the same sample, we test if the other bibliometric measures follow the known distribution laws