Linguistic summaries have been studied for many years and allow to sum up large volumes of data in a very intuitive manner. They have been studied over several types of data. However, few works have been led on graph databases. Graph databases are becoming popular tools and have recently gained significant recognition with the emergence of the so-called NoSQL graph databases. These databases allow users to handle huge volumes of data (e.g., scientific data, social networks). There are several ways to consider graph summaries. In this paper, we detail the specificities of NoSQL graph databases and we discuss how to summarize them by introducing several types of linguistic summaries, namely structure summaries, data structure summaries and fuzzy summaries. We present extraction methods that have been tested over synthetic and real database experimentations.