Identifying Hubs of Military Arms Export using Uniform Cost Search on Open-Source Information

Keywords
No Thumbnail Available
Date
2018-06-18
Language
en
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The purpose of this project is to identify military arms transport hubs in order to help to prevent arms reaching locations that are not stated in licenses that get released by governments. With the use of AI mechanisms, such as Dijkstra’s algorithm, that are able to make predictions about stops on the export route of military arms using open-source information, most inuential hubs can be spotted. The subsequent phase of the project then puts this information in human-usable notation, which ensures that it can be used in appropriate contexts. The gathered information aims to lower the nescience concerning weapon export. In a pipeline of 4 separate projects, an attempt is made to tackle the problem of losing track of military arms. Creating a web crawler that searches for appropriate information in the enormous amount of information the internet provides, is the first step of the pipeline. This step can be followed up the creation of a convolutional neural network that identifies military objects on images. After this, text on images can be recognized, which in the case of the pipeline is done for head stamps on military ammunition. The final step of the pipeline would be to use the gathered information to track and trace the routes the military products take to get from origin to destination and to find out what transport hubs are most important in these routes. This last part is discussed in this thesis. By using Dijkstra’s algorithm on open-source datasets about (Dutch) military export and about flights and airports all over the world, the shortest path can be found between a weapon’s origin country and its destination country. This is done by simulating flights between airports. Resulting from these shortest paths, some predictions for influential hubs in military transport can be formed. Even though the used data is limited because of monotonous and/or lacking information, predictions of important hubs in military transport can be deduced from the routes. These seem to be, among others, München (Germany), Girona (Spain) and Copenhagen (Denmark). For all of these found hubs, an explanation is given for why the hub might be indicated as important on military field. The process of this work and its results stress the importance of more clarity on military export and provide a basis for the further use of AI techniques in military areas outside of the battlefield.
Description
Citation
Faculty
Faculteit der Sociale Wetenschappen