Loading signal...
Signal
Submitted
Aug 15, 2026, 14:04 UTC
Large-scale hybrid human-AI analysis of 22 million global news titles reveals evolving media narratives on Russia-Ukraine from 2013 to 2024, highlighting coverage spikes around key events, dominant political figures, and Russian media language patterns.
The article 'Tracking the narrative: A data-driven analysis of media coverage of Russia and Ukraine 2013–2024' presents a comprehensive study using a hybrid human-AI approach to analyze 22.2 million news article titles from 14,622 sources across 194 countries and up to 74 languages, collected via Europe Media Monitor (EMM) from January 2013 to December 2024. The study investigates evolving media narratives concerning Russia and Ukraine over this period, focusing on key geopolitical events such as the Euromaidan protests, annexation of Crimea, and Russia's full-scale invasion of Ukraine in 2022. It uses multilingual clustering, cluster linking, and manual annotation to identify recurring narrative patterns, semantic similarity clusters, and dominant named entities such as Vladimir Putin, Volodymyr Zelenskyy, and Donald Trump. The analysis includes media coverage evolution, entity prominence dynamics, and language distribution of Russian-linked media, with sustained foreign language publication and varying conflict contextualizations. The article provides figures, tables, and detailed methodology on dataset construction, clustering using LaBSE embeddings, summarization via GPT-4 models, cluster linking with PaCMAP reduction, and human annotation processes. It emphasizes integrating framing theory, mediatisation theory, and strategic narrative theory within a scalable empirical framework to bridge theoretical gaps. The data and outputs are publicly available in the JRC Data Catalogue, with full open access under a Creative Commons Attribution License and no competing interests declared.
Confidence: High
The report details a comprehensive, empirical analysis of media narratives using computational methods and human validation to map information operations related to Russia and Ukraine without direct indication of a specific hostile act but showing patterns of propaganda and information influence campaigns over an extended period.
Source URL
https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0351627
Published as a hybrid-warfare signal with a public source and reviewer-approved assessment bands. See the methodology for how records are assessed.
{
"id": "8ec0bacf-4e95-4cb5-80b1-0d06d0ddfc47",
"url": "https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0351627",
"source": "extension",
"status": "approved",
"category": "Cyber & Information Warfare",
"subcategory": "Information Operations",
"incident_type": "Propaganda & disinformation campaigns",
"country": "Russian Federation",
"source_reliability": "C",
"info_credibility": "3",
"tags": [
"russia–ukraine war",
"media analysis",
"hybrid human-AI",
"multilingual clustering",
"strategic narratives",
"disinformation",
"digital propaganda",
"information operations",
"media monitoring",
"geopolitical events",
"named entity analysis"
],
"nodes": [],
"countries": [
"Russian Federation"
],
"where_text": "Cyberspace",
"headline": null,
"description": "The article 'Tracking the narrative: A data-driven analysis of media coverage of Russia and Ukraine 2013–2024' presents a comprehensive study using a hybrid human-AI approach to analyze 22.2 million news article titles from 14,622 sources across 194 countries and up to 74 languages, collected via Europe Media Monitor (EMM) from January 2013 to December 2024. The study investigates evolving media narratives concerning Russia and Ukraine over this period, focusing on key geopolitical events such as the Euromaidan protests, annexation of Crimea, and Russia's full-scale invasion of Ukraine in 2022. It uses multilingual clustering, cluster linking, and manual annotation to identify recurring narrative patterns, semantic similarity clusters, and dominant named entities such as Vladimir Putin, Volodymyr Zelenskyy, and Donald Trump. The analysis includes media coverage evolution, entity prominence dynamics, and language distribution of Russian-linked media, with sustained foreign language publication and varying conflict contextualizations. The article provides figures, tables, and detailed methodology on dataset construction, clustering using LaBSE embeddings, summarization via GPT-4 models, cluster linking with PaCMAP reduction, and human annotation processes. It emphasizes integrating framing theory, mediatisation theory, and strategic narrative theory within a scalable empirical framework to bridge theoretical gaps. The data and outputs are publicly available in the JRC Data Catalogue, with full open access under a Creative Commons Attribution License and no competing interests declared.",
"summary_short": "Large-scale hybrid human-AI analysis of 22 million global news titles reveals evolving media narratives on Russia-Ukraine from 2013 to 2024, highlighting coverage spikes around key events, dominant political figures, and Russian media language patterns.",
"summary_extended": null,
"actor_primary": "Russian media outlets",
"target_primary": null,
"location_text": "Russian Federation",
"location_region": null,
"location_coordinates_lat": 61.52401,
"location_coordinates_lon": 105.318756,
"location_precision": "Regional",
"is_aggressor_signal": false,
"signal_type": "HYBRID_SIGNAL",
"is_hybrid_war_signal": true,
"classification_complete": true,
"hybrid_signal_complete": true,
"classification_confidence": "High",
"classification_rationale": "The report details a comprehensive, empirical analysis of media narratives using computational methods and human validation to map information operations related to Russia and Ukraine without direct indication of a specific hostile act but showing patterns of propaganda and information influence campaigns over an extended period.",
"event_time_utc": "2025-07-24T12:00:00.000Z",
"end_time_utc": null,
"event_time_confidence": "unknown",
"created_at": "2026-08-15T14:04:00.868Z",
"updated_at": "2026-11-08T17:18:13.085Z",
"reliability_band": "strong",
"location_precision_band": "regional",
"time_precision_band": "unknown",
"severity_band": "moderate",
"attribution_confidence_band": "high",
"attribution_basis": "single_source"
}Source reliability
C
Info credibility
3
Event time
Jul 24, 2025, 12:00 UTC
Event time confidence
unknown
Location
Russian Federation
Primary actor
Russian media outlets
Country
Russian Federation
Countries
Russian Federation
Tags
russia–ukraine war, media analysis, hybrid human-AI, multilingual clustering, strategic narratives, disinformation, digital propaganda, information operations, media monitoring, geopolitical events, named entity analysis
Signal
Submitted
Aug 15, 2026, 14:04 UTC
Large-scale hybrid human-AI analysis of 22 million global news titles reveals evolving media narratives on Russia-Ukraine from 2013 to 2024, highlighting coverage spikes around key events, dominant political figures, and Russian media language patterns.
The article 'Tracking the narrative: A data-driven analysis of media coverage of Russia and Ukraine 2013–2024' presents a comprehensive study using a hybrid human-AI approach to analyze 22.2 million news article titles from 14,622 sources across 194 countries and up to 74 languages, collected via Europe Media Monitor (EMM) from January 2013 to December 2024. The study investigates evolving media narratives concerning Russia and Ukraine over this period, focusing on key geopolitical events such as the Euromaidan protests, annexation of Crimea, and Russia's full-scale invasion of Ukraine in 2022. It uses multilingual clustering, cluster linking, and manual annotation to identify recurring narrative patterns, semantic similarity clusters, and dominant named entities such as Vladimir Putin, Volodymyr Zelenskyy, and Donald Trump. The analysis includes media coverage evolution, entity prominence dynamics, and language distribution of Russian-linked media, with sustained foreign language publication and varying conflict contextualizations. The article provides figures, tables, and detailed methodology on dataset construction, clustering using LaBSE embeddings, summarization via GPT-4 models, cluster linking with PaCMAP reduction, and human annotation processes. It emphasizes integrating framing theory, mediatisation theory, and strategic narrative theory within a scalable empirical framework to bridge theoretical gaps. The data and outputs are publicly available in the JRC Data Catalogue, with full open access under a Creative Commons Attribution License and no competing interests declared.
Confidence: High
The report details a comprehensive, empirical analysis of media narratives using computational methods and human validation to map information operations related to Russia and Ukraine without direct indication of a specific hostile act but showing patterns of propaganda and information influence campaigns over an extended period.
Source URL
https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0351627
Published as a hybrid-warfare signal with a public source and reviewer-approved assessment bands. See the methodology for how records are assessed.
{
"id": "8ec0bacf-4e95-4cb5-80b1-0d06d0ddfc47",
"url": "https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0351627",
"source": "extension",
"status": "approved",
"category": "Cyber & Information Warfare",
"subcategory": "Information Operations",
"incident_type": "Propaganda & disinformation campaigns",
"country": "Russian Federation",
"source_reliability": "C",
"info_credibility": "3",
"tags": [
"russia–ukraine war",
"media analysis",
"hybrid human-AI",
"multilingual clustering",
"strategic narratives",
"disinformation",
"digital propaganda",
"information operations",
"media monitoring",
"geopolitical events",
"named entity analysis"
],
"nodes": [],
"countries": [
"Russian Federation"
],
"where_text": "Cyberspace",
"headline": null,
"description": "The article 'Tracking the narrative: A data-driven analysis of media coverage of Russia and Ukraine 2013–2024' presents a comprehensive study using a hybrid human-AI approach to analyze 22.2 million news article titles from 14,622 sources across 194 countries and up to 74 languages, collected via Europe Media Monitor (EMM) from January 2013 to December 2024. The study investigates evolving media narratives concerning Russia and Ukraine over this period, focusing on key geopolitical events such as the Euromaidan protests, annexation of Crimea, and Russia's full-scale invasion of Ukraine in 2022. It uses multilingual clustering, cluster linking, and manual annotation to identify recurring narrative patterns, semantic similarity clusters, and dominant named entities such as Vladimir Putin, Volodymyr Zelenskyy, and Donald Trump. The analysis includes media coverage evolution, entity prominence dynamics, and language distribution of Russian-linked media, with sustained foreign language publication and varying conflict contextualizations. The article provides figures, tables, and detailed methodology on dataset construction, clustering using LaBSE embeddings, summarization via GPT-4 models, cluster linking with PaCMAP reduction, and human annotation processes. It emphasizes integrating framing theory, mediatisation theory, and strategic narrative theory within a scalable empirical framework to bridge theoretical gaps. The data and outputs are publicly available in the JRC Data Catalogue, with full open access under a Creative Commons Attribution License and no competing interests declared.",
"summary_short": "Large-scale hybrid human-AI analysis of 22 million global news titles reveals evolving media narratives on Russia-Ukraine from 2013 to 2024, highlighting coverage spikes around key events, dominant political figures, and Russian media language patterns.",
"summary_extended": null,
"actor_primary": "Russian media outlets",
"target_primary": null,
"location_text": "Russian Federation",
"location_region": null,
"location_coordinates_lat": 61.52401,
"location_coordinates_lon": 105.318756,
"location_precision": "Regional",
"is_aggressor_signal": false,
"signal_type": "HYBRID_SIGNAL",
"is_hybrid_war_signal": true,
"classification_complete": true,
"hybrid_signal_complete": true,
"classification_confidence": "High",
"classification_rationale": "The report details a comprehensive, empirical analysis of media narratives using computational methods and human validation to map information operations related to Russia and Ukraine without direct indication of a specific hostile act but showing patterns of propaganda and information influence campaigns over an extended period.",
"event_time_utc": "2025-07-24T12:00:00.000Z",
"end_time_utc": null,
"event_time_confidence": "unknown",
"created_at": "2026-08-15T14:04:00.868Z",
"updated_at": "2026-11-08T17:18:13.085Z",
"reliability_band": "strong",
"location_precision_band": "regional",
"time_precision_band": "unknown",
"severity_band": "moderate",
"attribution_confidence_band": "high",
"attribution_basis": "single_source"
}Source reliability
C
Info credibility
3
Event time
Jul 24, 2025, 12:00 UTC
Event time confidence
unknown
Location
Russian Federation
Primary actor
Russian media outlets
Country
Russian Federation
Countries
Russian Federation
Tags
russia–ukraine war, media analysis, hybrid human-AI, multilingual clustering, strategic narratives, disinformation, digital propaganda, information operations, media monitoring, geopolitical events, named entity analysis