Detection of online radicalisation in the Maldivian language, Dhivehi

Author Identifier

Hussain Ibrahim

Date of Award


Document Type



Edith Cowan University

Degree Name

Doctor of Philosophy (Integrated)


School of Science

First Supervisor

Ahmed Ibrahim

Second Supervisor

Mike Johnstone

Third Supervisor

Shehenaz Adam


Early detection of online radical content is important for intelligence services to combat radicalisation and terrorism. This research was motivated by the lack of language tools in the detection of radicalisation in Dhivehi. The radical text data that is populated online in Dhivehi is increasing and taxing the capabilities of human open-source intelligence operatives. This research answered the question, “What methods are most suited for the detection of online radicalisation in the Maldivian language, Dhivehi?”. At the time of this research, there were no known language-based technologies or datasets for Dhivehi that can assist in understanding the syntax and semantics of the language to detect online radicalisation.

This research applied Machine Learning and Natural Language Processing to detect online radicalisation content in Dhivehi, with the incorporation of domain-specific knowledge from the Maldives. As a result, it has produced datasets that would be beneficial for further research. The research used Machine Learning to evaluate the most effective technique for detection of radicalisation text in Dhivehi and used interviews with subject matter experts and self-deradicalised individuals to both validate the results, add contextual information and improve recognition accuracy.

The research found that the Naïve Bayes algorithm worked best for the detection of radicalisation text in Dhivehi. The interviews with the self-deradicalised individuals found that online radicalisation played an important role in the radicalisation of Maldivians. However, the online content followed by the interviewees were mainly in English.

Significant and original contributions of this research to the existing body of knowledge includes datasets in the form of labelled radical/non-radical text, sentiment corpus of radical words and primary interview data of self-deradicalised individuals and a technique for detection of radicalisation text in Dhivehi for the first-time using Machine Learning.



Access Note

Access to this thesis is embargoed until 18th October 2026.

Access to this thesis is restricted. Please see the Access Note below for access details.