Open the European Security Monitor

ABOUT THE MONITOR

Understanding the conversation around European security.

The European Security Monitor is an independent research project created and developed by Kedma Hamelberg, PhD. It explores how public videos discuss European security, sovereignty and essential resources, and how selected comments respond.

What you are seeing

The monitor presents a weekly sample of public material. AI-assisted analysis uses available source text, including translations and audio transcriptions, alongside human review. Published AI labels are provisional; human corrections take precedence. Sources without usable evidence remain unclassified.

The map connects countries mentioned in the content with video engagement and audience responses. Agreement with a message and the emotional tone of a comment are different measures. Detailed panels retain uncertainty rather than forcing every response into a positive or negative category.

How to read the map

Sizes are relative within the selected measure; small symbols have a minimum readable size. Only classified videos appear; green circles include related and unrelated classified sources. Country mentions locate the content, not the audience. Select any symbol for a country briefing. All retained weeks use the latest snapshot per video, not a sum of lifetime counters.

How to read the results

These observations describe the retained sample, not national public opinion or a real-time threat assessment. Coverage is incomplete and uneven. Translation, speech recognition and AI classification can introduce errors. Describing an allegation does not verify it.

Country markers identify content mentions, not viewers’ locations. Views are lifetime platform counts, not unique people. Country percentages may overlap when a video mentions several countries. Time-window and filter labels explain which observations are included; a single snapshot cannot establish a trend or causation.

Raw transcripts, comment text and personal identifiers are not published here. The public display uses source links and aggregate results. The approach draws on communication research, including stance detection and an exploratory adaptation of message-delivery research; it is not a validated measure of influence or effectiveness.

Message-delivery categories distinguish evidence found in the available text, a feature not observed in that text, and a feature that cannot be assessed from the available evidence. Not observed in text is not proof of absence from the full audiovisual video. Confirmed absence requires human viewing. Categories can overlap.

Explore Kedma’s work