Inferring itineraries of containerized cargo through the application of Conditional Random Fields
Pedro Chahuara, Luca Mazzola, Michail Makridis, Claudio Schifanella, Aris Tsois, Mauro Pedone · 2014
This paper proposes a method to infer the itinerary of cargo transported in shipping containers based on a large, heterogeneous and noisy dataset of Container Status Messages. Such itinerary information can be used to improve the risk analysis performed by authorities in their effort to secure the global trade and fight frauds. Our method, based on conditional random fields, is able not only to partition the original noisy dataset into appropriate sequences describing distinct shipments of containerized cargo but also to identify the messages that describe the various stages of the transportation. The experiments performed suggest that conditional random fields provide a high accuracy for this sequential pattern mining problem.