{"url":"/dataset/gdelt","name":"GDELT","full_name":null,"description_markdown":"The **GDELT Project** is a remarkable initiative that monitors our world by analyzing global news from various sources. Here are the key aspects of the GDELT dataset:\r\n\r\n1. **Scope and Purpose**:\r\n   - The GDELT Project aims to create a comprehensive, **real-time database** of global human society.\r\n   - It monitors news from **broadcasts, print media, and web sources** in nearly every country and over **100 languages**.\r\n   - By analyzing this vast dataset, it identifies people, locations, organizations, themes, emotions, and events that shape our global society every second of every day.\r\n\r\n2. **Data Collection**:\r\n   - GDELT continuously captures and analyzes news articles, broadcasts, and online sources.\r\n   - Its historical archives date back to **January 1, 1979**, and it updates every **15 minutes**.\r\n   - The project goes beyond Western media, providing a more **global perspective** on world events and sentiments.\r\n\r\n3. **Features**:\r\n   - GDELT uses sophisticated natural language and data mining algorithms, including powerful deep learning techniques.\r\n   - It extracts over **300 categories of events**, millions of themes, thousands of emotions, and the networks connecting them.\r\n   - The dataset models human interactions at a large scale, making it valuable for research and analysis.\r\n\r\n4. **Vision**:\r\n   - The GDELT Project envisions using this data to:\r\n     - Understand the world through others' eyes.\r\n     - Break down language and access barriers.\r\n     - Facilitate conversations between societies.\r\n     - Empower local populations with information for safer lives.\r\n     - Map happiness, conflict, and potentially forecast global tensions.\r\n\r\n5. **Global Reach**:\r\n   - GDELT monitors media in **over 100 languages** across every country, providing a truly global perspective.\r\n   - It allows us to explore how social media is used worldwide and how people express themselves online.\r\n\r\n6. **Open Data**:\r\n   - The entire GDELT database is **free and open**.\r\n   - Researchers can download raw data, visualize it, or analyze it at scale using tools like **Google BigQuery**¹²³⁴⁵.\r\n\r\nSource: Conversation with Bing, 3/12/2024\r\n(1) The GDELT Project. https://www.gdeltproject.org/.\r\n(2) The GDELT Database | Aalto Datahub. https://datahub.aalto.fi/en/data-sources/the-gdelt-database.\r\n(3) An Introduction to GDELT Data | MongoDB. https://www.mongodb.com/developer/products/mongodb/introduction-to-gdelt-data/.\r\n(4) GDELT 2.0: Our Global World in Realtime – The GDELT Project. https://blog.gdeltproject.org/gdelt-2-0-our-global-world-in-realtime/.\r\n(5) Data: Querying, Analyzing and Downloading: The GDELT Project. https://www.gdeltproject.org/data.html.","description_withheld":null,"homepage":"https://www.gdeltproject.org/","introduced_date":null,"introduced_date_note":null,"introduced_by":null,"license":null,"modalities":[],"tasks":[{"name":"Link Prediction","url":"/task/link-prediction","datasets_with_task":"/datasets/task/link-prediction"}],"languages":[],"variants":["GDELT"],"data_loaders":[],"num_papers_in_archive":3,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/link-prediction-on-gdelt","task":"Link Prediction","dataset_variant":"GDELT","rows":11,"metrics":["MRR"],"first_row_in_archive_order":{"model":"SPA","paper":"/paper/search-to-pass-messages-for-temporal","metrics":{"MRR":"0.36"},"code_links":[{"title":"striderdu/spa","url":"https://github.com/striderdu/spa"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/search-to-pass-messages-for-temporal","title":"Search to Pass Messages for Temporal Knowledge Graph Completion","date":"2022-10-30","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/along-the-time-timeline-traced-embedding-for","title":"Along the Time: Timeline-traced Embedding for Temporal Knowledge Graph Completion","date":"2022-10-17","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/rotateqvs-representing-temporal-information","title":"RotateQVS: Representing Temporal Information as Rotations in Quaternion Vector Space for Temporal Knowledge Graph Completion","date":"2022-03-15","rows_on_this_dataset":9,"code_links":0,"syntology":null}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":0,"samples_harvested":0,"samples_ran":0,"samples_unverified":0,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}