We use cookies to ensure that we give you the best experience on our website. By continuing to browse this repository, you give consent for essential cookies to be used. You can read more about our Privacy and Cookie Policy.

Durham e-Theses
You are in:

Towards a National Security Analysis Approach via Machine
Learning and Social Media Analytics

CARDENAS-CANTO, PEDRO (2022) Towards a National Security Analysis Approach via Machine
Learning and Social Media Analytics.
Doctoral thesis, Durham University.

[img]PDF (Accepted Version of the thesis) - Accepted Version


Various severe threats at national and international level, such as health crises, radicalisation, or organised crime, have the potential of unbalancing a nation's stability. Such threats impact directly on elements linked to people's security, known in the literature as human security components. Protecting the citizens from such risks is the primary objective of the various organisations that have as their main objective the protection of the legitimacy, stability and security of the state.

Given the importance of maintaining security and stability, governments across the globe have been developing a variety of strategies to diminish or negate the devastating effects of the aforementioned threats. Technological progress plays a pivotal role in the evolution of these strategies. Most recently, artificial intelligence has enabled the examination of large volumes of data and the creation of bespoke analytical tools that are able to perform complex tasks towards the analysis of multiple scenarios, tasks that would usually require significant amounts of human resources.

Several research projects have already proposed and studied the use of artificial intelligence to analyse crucial problems that impact national security components, such as violence or ideology. However, the focus of all this prior research was examining isolated components. However, understanding national security issues requires studying and analysing a multitude of closely interrelated elements and constructing a holistic view of the problem.

The work documented in this thesis aims at filling this gap. Its main contribution is the creation of a complete pipeline for constructing a big picture that helps understand national security problems. The proposed pipeline covers different stages and begins with the analysis of the unfolding event, which produces timely detection points that indicate that society might head toward a disruptive situation. Then, a further examination based on machine learning techniques enables the interpretation of an already confirmed crisis in terms of high-level national security concepts.

Apart from using widely accepted national security theoretical constructions developed over years of social and political research, the second pillar of the approach is the modern computational paradigms, especially machine learning and its applications in natural language processing.

Item Type:Thesis (Doctoral)
Award:Doctor of Philosophy
Faculty and Department:Faculty of Science > Computer Science, Department of
Thesis Date:2022
Copyright:Copyright of this thesis is held by the author
Deposited On:10 Jan 2023 11:35

Social bookmarking: del.icio.usConnoteaBibSonomyCiteULikeFacebookTwitter