This paper propose a method to infer the itinerary followed by goods transported in shipping containers.
The final application can have a positive impact in risk analysis of transported cargo performed by authorities.
Our work relies on Container Status Message as source of information, which is often characterized to contain ambiguous, incomplete, imprecise and redundant data since it is generated from heterogenous sources.
We propose a method, based on conditional random fields, in order to obtain sequences of messages characterizing containerized goods trajectories from large and noisy data sets.
The experiments, which were performed with a set of container status messages from different carriers, suggest that CRF provides a high accuracy for estimating the cargo itineraries.
CHAHUARA QUISPE Pedro;
MAZZOLA Luca;
MAKRIDIS Michail;
SCHIFANELLA Claudio;
TSOIS Aris;
PEDONE Mauro;
2015-01-05
The Institute of Electrical and Electronics Engineers, Inc
JRC90282
978-1-4799-6363-8,
http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6975565,
https://publications.jrc.ec.europa.eu/repository/handle/JRC90282,
10.1109/JISIC.2014.29,
| Name | Country | City | Type |
|---|
This document is only visible at the Commission level.
You are not authorized to publish or distribute it outside the European Commission.
This is a public document. You can share this publication.
Datasets
| ID | Title | Public URL |
|---|
Dataset collections
| ID | Acronym | Title | Public URL |
|---|
Scripts / source codes
| Description | Public URL |
|---|
Additional supporting files
| File name | Description | File type |
|---|