The Automatic Identification System (AIS) is an electronic system set on board vessels, and transmits its location, amongst many other data. As messages are sent and received by vessel and coastal stations within the radio horizon, it enables a better understanding of the surroundings for vessel and coastal states. However, this system has weaknesses, as errors in data and falsification such as identity theft or disappearances have been demonstrated. This paper presents a methodology based on the notion of data integrity, in which data coming from AIS is processed in different ways in order to produce confidence coefficient on the genuineness and truthfulness of the data within. Such a study is done using an architecture built around a relational database using geospatial predicates. As AIS falsification brings maritime risks, a geospatial analysis of risks is proposed, based on both the AIS integrity assessment and geographic features around the studied vessel, leading to a better response from competent authorities in case of need, an overall improvement of maritime situational awareness and an enhancement of the safety and security of maritime navigation.