CoolFace
Datasetpublic

fahadh4ilyas/indonesian_news_datasets

Indonesian News Datasets This dataset is a modification from dataset repo iqballx/indonesian_news_datasets. Column label is added with these categories: POLITIK_PEMERINTAHAN EKONOMI_BISNIS HUKUM_KRIMINAL OLAHRAGA TEKNOLOGI_DIGITAL BENCANA_LINGKUNGAN The label is created using prompt to LLM Qwen/Qwen3-30B-A3B-Instruct-2507-FP8. Disclaimer: The resulted label has not been currated and just received as is from LLM. The dataset compiles information from seven prominent… See the full description on the dataset page: https://huggingface.co/datasets/fahadh4ilyas/indonesian_news_datasets.

sourceHugging Facecc-by-nc-4.0updated 7mo agoView on Hugging Face
0likes33downloads
Dataset Card

Indonesian News Datasets

This dataset is a modification from dataset repo iqballx/indonesian_news_datasets. Column label is added with these categories:

  • —POLITIK_PEMERINTAHAN
  • —EKONOMI_BISNIS
  • —HUKUM_KRIMINAL
  • —OLAHRAGA
  • —TEKNOLOGI_DIGITAL
  • —BENCANA_LINGKUNGAN

The label is created using prompt to LLM Qwen/Qwen3-30B-A3B-Instruct-2507-FP8.

Disclaimer: The resulted label has not been currated and just received as is from LLM.

The dataset compiles information from seven prominent Indonesian news platforms: Tempo, CNN Indonesia, CNBC Indonesia, Okezone, Suara, Kumparan, and JawaPos. Each source contributes a diverse range of articles, collectively forming a comprehensive repository of Indonesian news content. The dataset consists of diverse columns: 'id' denotes the unique identifier for each news article, 'title' contains the headline, 'image' refers to the associated visual content, 'url' specifies the source link, 'content' encompasses the full text of the news article, 'embedding' houses the text embeddings extracted using the OpenAI text-embedding-ada-002 model, 'createdat' marks the creation timestamp, 'updatedat' signifies the last update timestamp, and 'summary' encapsulates the concise article summary generated via the ChatGPT API. These columns collectively represent a comprehensive set of information for each news article, facilitating various natural language processing tasks and analysis while providing enriched contextual details for effective exploration and modeling.